Summary of External Peer Review and Public Comments and Disposition for 1,4-Dioxane: Response to Support Risk Evaluation of 1,4-Dioxane December 2020


a rnjt

States

Environmental Protection Agency

Office of Chemical Safety and
Pollution Prevention
December 2020

Summary of External Peer Review and Public Comments and Disposition

for 1,4-Dioxane:

Response to Support Risk Evaluation of 1,4-Dioxane

December 2020

-------
Table of Contents

Abbreviations	4

1.	Content and Organization	11

2.	Systematic Review	29

3.	Environmental Fate, Exposure & Effects	38

4.	Exposure and Releases	73

5.	Human Health	102

6.	Risk Characterization	129

7.	Editorial Comments	156

8.	Supplemental Analysis	158

Page 2 of212

-------
This document summarizes the public and external peer review comments that the EPA's Office of Pollution Prevention and Toxics
(OPPT) responses to the comments received for the draft risk evaluation. This document also summarizes the public comments and
EPA/OPPT's responses to the comments received for the draft supplemental analysis to the draft risk evaluation of 1,4-dioxane. It also
provides EPA/OPPT's response to the comments received from the public and the peer review panel.

EPA/OPPT appreciates the valuable input provided by the public and peer review panel. The input resulted in numerous revisions to
the risk evaluation document.

The peer review and public comments are categorized by the 1,4-dioxane peer review charge questions, which align with the seven
themes listed below (including the addition of a section for editorial comments). Within each theme, comments that cover similar
issues are presented together.

1.

Content and Organization

2.

Systematic Review

3.

Environmental Fate, Exposure & Effects

4.

Exposure and Releases

5.

Human Health

6.

Risk Characterization

7.

Editorial Comments

8.

Supplemental Analysis

Page 3 of212

-------
Abbreviations

ACC

American Chemistry Council

ACE

Acute-to-Chronic Estimation

ADC

Average daily concentration

AF

Assessment factor

AFL-CIO

American Federation of Labor and Congress of Industrial Organizations

AEGL

Acute Exposure Level Guidelines

AIHA

American Industrial Hygiene Association

AOP

Adverse outcome pathway

APF

Assigned protection factor

APHA

American Public Health Association

APHL

Association of Public Health Laboratories

AQMD

Air Quality Management District

AT SDR

Agency for Toxic Substances and Disease Registry

BCF

Bioconcentration Factor

BMDL

Benchmark dose lower bound

BMDS

Benchmark Dose Software

BW

Body weight

CAA

Clean Air Act

CalEPA

California Environmental Protection Agency

CASRN

Chemical Abstracts Service Registry Number

CEM

Consumer Exposure Model

COU

Condition of use

CFD

Computational fluid dynamics

ChV

Chronic value

CNS

Central Nervous System

coc

Concentration of concern

CWA

Clean Water Act

DMR

Discharge Monitoring Report

EC50

Effect Concentration at which 50% of test organisms exhibit the effect

ECEL

Existing Chemical Concentration Limit

EDF

Environmental Defense Fund

E-FAST

Exposure and Fate Assessment Screening Tool

Page 4 of212

-------
EIA

Environmental Investigation Agency

EPI Suite™

Estimation Programs Interface suite of models

EPN

Environmental Protection Network

EXAMS

Exposure Analysis Modeling System

GHS

Globally Harmonized System

HAP

Hazardous air pollutant

HEC

Human equivalent concentration

HEI

Health Effects Institute

HERO

Health & Environmental Research Online

HUC

Hydrologic unit code

I ARC

International Agency for Research on Cancer

IUR

Inhalation unit risk

Koa

Octanol-Air Partition Coefficient

Koc

Soil Organic Carbon-Water Partitioning Coefficient

LA DC

Lifetime average daily concentrations

LCoi

Lethal Concentration at which 1% of test organisms die

LCio

Lethal Concentration at which 10% of test organisms die

LC50

Lethal Concentration at which 50% of test organisms die

LOAEL

Lowest Observed Adverse Effect Level

LOD

Limit of detection

MOA

Mode of Action

MOE

Margin of Exposure

NAICS

North American Industry Classification System

NAS

National Academies of Science

NATA

National Air Toxics Assessment

NEI

National Emissions Inventory

NESHAP

National Emission Standards for Hazardous Air Pollutants

NF

Near-field

NHANES

National Health and Nutrition Examination Survey

NIOSH

National Institute for Occupational Safety and Health

NOAEL

No Observed Adverse Effect Level

NOEC

No Observed Effect Concentration

NPDES

National Pollutant Discharge Elimination System

Page 5 of212

-------
NTP

National Toxicology Program

OECD

Organisation for Economic Co-operation and Development

OES

Occupational exposure scenario

OPERA

Open Structure-activity/property Relationship App

OPPT

Office of Pollution Prevention and Toxics

ONU

Occupational non-user

OSHA

Occupational Safety and Health Administration

PBPK

Physiologically based pharmacokinetic

PESS

Potentially Exposed or Susceptible Subpopulations

PEL

Permissible exposure limits

PDM

Probabilistic Dilution Model

PF

Protection factor

POD

Point of departure

POTW

Publicly owned treatment works

PPE

Personal protective equipment

QSAR

Quantitative Structure-Activity Relationship

REL

Reference Exposure Level

RIOPA

Relationship between Indoor, Outdoor, and Personal Air

ROS

Regression on Order Statistics

RQ

Risk quotient

SACC

Science Advisory Committee on Chemicals

SCHF

Safer Chemicals Healthy Families

SDS

Safety Data Sheet

SIR

Standard incidence rates

SOCMA

Society of Chemical Manufacturers & Affiliates

STORET

STOrage and RETrieval database

TNO

The Netherlands Organisation for Applied Scientific Research

TRI

Toxics Release Inventory

TSCA

Toxic Substances Control Act

TURI

Toxics Use Reduction Institute

TWA

Time-weighted average

UCSF PRHE

University of California, San Francisco Program on Reproductive Health and the Environment

UF

Uncertainty factor

Page 6 of212

-------
U.S. BLS	United States Bureau of Labor Statistics

USGS	U.S. Geological Survey

WHO	World Health Organization

Page 7 of212

-------
List of Comments

Uisk Kvaluation

#

Docket l-'ilc

Submitter

14



(iiirv A Buchanan. Director. New Jersey Deparinienl of 1 ji\ironnienlal Protection
(NJDEP)

15

EP A-HG-GPPT-20

19-023 8-i

Liz Hitchcock, Acting Director, Safer Chemicals Healthy Families et al.

19

EP A-HG-GPPT-20

19-02

Environmental Protection Network (EPN)

20

EP A-HG-GPPT-20

19-0238-0020

Ben Gann, Director, Chemical Products & Technology Division, American Chemistry
Council's (ACC) North American Flame Retardant Alliance (NAFRA)



21

EP A-HO-OPPT-2019-023 8-0021

Jonathan Kalmuss-Katz, Staff Attorney, Earthjustice and Randy Rabinowitz,
Executive Director, Occupational Safety & Health (OSH) Law Project

22

EP A-HG-GPPT-2019-023 8-0022

Stephen P. Risotto, Senior Director, American Chemistry Council (ACC)

23

EP A-HO-OPPT-2019-023 8-0023

Suzanne Hartigan, Senior Director, Regulatory and Technical Affairs, American
Chemistry Council (ACC)

24

EP A-HO-OPPT-2019-023 8-C

Richard A. Denison, Lead Senior Scientist, on behalf of Environmental Defense Fund
(EDF)

30

EP A-HO-OPPT-2019-023 8-003 0

Douglas M. Troutman, Sr. Vice President, Government Affairs, American Cleaning
Institute (ACI) and Michael R. Gruber, Vice President, Government Affairs, Grocery
Manufacturers Association (GMA)

31

EP A-HG-GPPT-20

19-0238-0031

Steve Risotto, American Chemistry Council (ACC)

32

EP A-HO-OPPT-20

19-0238-0032

Huntington Breast Cancer Action Coalition, Inc (HBCAC)

40

EP A-HO-OPPT-20

19-0238-0040

Brett Howard, Director, Regulatory & Technical Affairs American Chemistry Council

42

EP A-HO-OPPT-20

19-0238-0042

Emily Sutton, Haw Riverkeeper, Haw River Assembly

43

EP A-HO-OPPT-20

19-0238-0043

Catherine Neuschler, Manager, Water Assessment Section, Environmental Analysis
and Outcomes Division, Minnesota Pollution Control Agency (MPCA) and Jim Kelly,
Manager, Environmental Surveillance & Assessment, Environmental Health Division,
Minnesota Department of Health (MDH)



44

EP A-HO-OPPT-2019-023 8-0044

James R. Fletchtner, Executive Director, Cape Fear Public Utility Authority (CFPUA)

45

EPA-H<

Mack McKinley, Water Resources Engineer, City of Sanford, Florida

Page 8 of212

-------
46

EP A-HO-OPPT-2019-023 8-C

Jonathan Kalmuss-Katz, Staff Attorney, Earthjustice and Randy Rabinowitz,
Executive Director, Occupational Safety & Health Law Project on behalf of American
Federation of Labor and Congress of Industrial Organizations et al.

47

EP A-HO-OPPT-2019-02:

Sonya Lunder, Senior Toxics Policy Advisor Gender, Equity & Environment
Program, Sierra Club

48

EP A-HO-OPPT-2019-023 8-0048

Suzanne Hartigan, Senior Director, Regulatory and Technical Affairs, American
Chemistry Council (ACC)

49

EP A-HO-OPPT-2019-023 8-0049

G. Tracy Mehan III, Executive Director, Government Affairs, American Water Works
Association (AWW A) and Diane VanDe Hei, Chief Executive Officer, Association of
Metropolitan Water Agencies (AMWA)

50

EP A-HO-OPPT-2019-023 8-0050

Amble Johnson, Jean Zhuang and Kelly Moser, Attorneys, Southern Environmental
Law Center (SELC)

51

EP A-HO-OPPT-20

19-0238-0051

Tosh Sagar, Senior Associate Attorney, Earthjustice

52

EP A-HO-OPPT-20

19-0238-0052

Brent Tracy, Senior Director and Associate General Counsel, Johns Manville (JM)

53

EP A-HO-OPPT-20

19-0238-0053

D. Peter

54

EP A-HO-OPPT-20

19-0238-0054

Amble Johnson, Jean Zhuang and Kelly Moser, Attorneys, Southern Environmental
Law Center (SELC)



55

EP A-HO-OPPT-2019-023 8-005 5

Liz Hitchcock, Acting Director, Safer Chemicals Healthy Families (SCHF) et al.

56

EP A-HO-OPPT-2019-023 8-0056

Swati Rayasam, Science Associate, Program on Reproductive Health and the
Environment, Department of Obstetrics, Gynecology and Reproductive Sciences,
University of California, San Francisco et al.

57

EP A-HO-OPPT-2019-023 8-005 7

Tosh Sagar, Senior Associate Attorney, Earthjustice on behalf of Achieving
Community Tasks Successfully (ACTS)

58

EP A-HO-OPPT-20

19-0238-0058

Stephanie Schwarz, Legal Fellow, Environmental Defense Fund (EDF)

59

EP A-HO-OPPT-20

19-0238-0059

Stephen P. Risotto, Senior Director, American Chemistry Council (ACC)

60

EP A-HO-OPPT-20

19-0238-0060

Cape Fear River Watch (Part 1 of 3)

61

EP A-HO-OPPT-20

19-0238-0061

Cape Fear River Watch (Part 2 of 3)

62

EP A-HG-GPPT-20

19-0238-0062

Cape Fear River Watch (Part 3 of 3)

SACC

N/A

Science Advisory Committee on Chemicals (SACC)



Page 9 of212

-------
Supplemental Analysis

75



.1 Akin Rohei'son. l-\eaili\e Director. Association of Stale Drinking Water
Administrators (ASDWA)

76

EP A-HO-OPPT-2019-02:

Diane VanDe Hei, Chief Executive Officer, Association of Metropolitan Water
Agencies (AMWA)

77

EP A-HG-GPPT-20

19-0238-0077

Elizabeth Hitchcock, Safer Chemicals Healthy Families et al.

78

EP A-HG-GPPT-20

19-0238-0078

Michelle Roos, Environmental Protection Network (EPN)

79

EP A-HO-OPPT-20

19-0238-0079

Stephen Wieroniey, Director, American Chemistry Council's (ACC) Spray Foam
Coalition (SFC)

80

EP A-HO-OPPT-2019-023 8-0080

Stephen P. Risotto, Senior Director, American Chemistry Council (ACC)

81

EPA-H<

G. Tracy Mehan, III, Executive Director - Government Affairs, American Water
Works Association (AWWA)

82

EP A-HO-OPPT-2019-023 8-0082

Vincent Cogliano, Deputy Director for Scientific Programs, California Office of
Environmental Health Hazard Assessment

83

EP A-HG-GPPT-2019-023 8-0083

Richard Denison, Lead Senior Scientist, Environmental Defense Fund (EDF)

84

EP A-HO-OPPT-2019-023 8-0084

Kathleen Stanton, Associate Vice President, Technical & International Affairs,
American Cleaning Institute (ACI) and Steven Bennett, Senior Vice President,
Scientific & Regulatory Affairs, Household and Commercial Products Association
(HCPA)

85

EP A-HO-OPPT-2019-023 8-0085

Letitia James, Attorney General of New York, Sarah Kam, Assistant Attorney
General, Office of the Attorney General, New York State et al.

86

EP A-HG-GPPT-2019-023 8-0086

Meredith Williams, Director, Department of Toxic Substances Control, California
Department of Toxic Substances Control (DTSC)

87

EP A-HO-OPPT-2019-023 8-0087

J. Alan Roberson, Executive Director, Association of State Drinking Water
Administrators (ASDWA)

88

EP A-HO-OPPT-2019-023 8-0088

Jonathan Kalmuss-Katz and Rashmi Joglekar, Earthjustice, and Randy Rabinowitz,
Executive Director, Occupational Safety & Health Law Project on behalf of American
Federation of Labor and Congress of Industrial Organizations (AFL-CIO) et al.

89

EP A-HO-OPPT-2019-023 8-0089

Elizabeth Hitchcock, Safer Chemicals Healthy Families et al.

90

EP A-HG-GPPT-2019-023 8-0090

Suzanne Hartigan, Senior Director, Regulatory and Technical Affairs, American
Chemistry Council (ACC)

Page 10 of 212

-------
1. Content and Organization

Charge Question 1.1: Please comment on the overall content, organization, and presentation of the draft risk evaluation of 1,4-dioxane.
Charge Question 1.2: Please provide suggestions for improving the clarity of the information presented in the documents.

Summary of Comments lor Specific
Issues Related to Charge Question 1

I.IW/OPPT Response

(ieneral Comments

SACC

• Provide a brief history and basis for why the chemical is under risk
evaluation. While much of this information is introduced in the
Scope and Problem Formulation, inclusion in the Evaluation would
greatly enhance the final product.

• Improve the clarity of the risk evaluation with careful review,
editing, and inclusion of additional graphics. All section references
(including appendices and their subheadings) should be formatted
as hyperlinks to support easier review and reading.

• EPA should provide definitions for all specific terms early in the
main document, not just by giving a citation to other document.
This could be done by incorporating a glossary of terms.

• A basis for the chemical risk evaluation is
provided in the Introduction (Section 1).

• Each of the recommendations to improve
clarity within the risk evaluation have been
accepted.

• While EPA has not included a glossary,
definitions have been provided for terms
within the context of their use in the
document.

Information Authorities

• Why did EPA fail to use its information authorities under TSCA to
require submission and/or development of relevant information to
fill numerous data gaps, including an absence of sufficient
environmental monitoring data; environmental fate data;
ecotoxicity data; product/use and concentration data; inhalation
exposure data; dermal exposure data; dermal toxicity data; and
reproductive/developmental/ neurodevelopmental toxicity data?

EPA had sufficient information to complete the
1,4-dioxane risk evaluation using a weight of
scientific evidence approach. EPA selected the
first 10 chemicals for risk evaluation based in
part on its assessment that these chemicals
could be assessed without the need for
regulatory information collection or
development. When preparing this risk
evaluation, EPA obtained and considered
reasonably available information, defined as
information that EPA possesses, or can
reasonably obtain and synthesize for use in risk

Page 11 of 212

-------
evaluations, considering the deadlines for
completing the evaluation. In some cases,
when information available to EPA was
limited, the Agency relied on models; the use of
modeled data is in line with EPA's final Risk
Evaluation Rule and EPA's risk assessment
guidelines.

As further noted in the response to the
comments on the scope documents, EPA
conducted extensive and varied data gathering
activities for each of the first 10 chemicals,
including:

Extensive and transparent searches of
public databases and sources of scientific
literature, government and industry sector
or other reports;

Searches of EPA TSC A section 8(e)
Substantial Risk Reporting, Chemical Data
Reporting, and Toxics Release Inventory
databases and other EPA information
holdings; and CBI submission holdings;

Searches for Safety Data Sheets (SDSs)
using the internet, EPA Chemical and
Product Categories (CPCat) data, the
National Institute for Health's (NIH)
Household Product Database, and other
resources in which SDS could be found;

Preparation of a market analysis using
proprietary databases and repositories;

Outreach meetings with chemical
manufacturers, processors, chemical users,
non-governmental organizations, trade	

Page 12 of 212

-------




organizations, and other experts, including
other State and Federal Agencies (e.g.,
Department of Defense, NASA, OSHA,
NIOSH, FDA and CPSC); and

Publication of conditions of use
documents, scope documents, and problem
formulation documents to solicit
information generally from industry,
nongovernmental organizations, and the
public.

24, 46,
48, 50,
54, 58

Best Available Science

• In several instances, EPA failed to utilize the "best available

science" "without consideration of costs or other non-risk factors"
and all "reasonably available information" (15 U.S.C. § 2625(h)).
o For example, EPA relied on outdated TRI data, choosing to
use data from 2015, even though data from 2016 and 2017
are readily available; the rationale that incorporation of
more recent reporting years was not expected to alter
conclusions of the screening-level assessment is not
acceptable.

• At the time of original project scoping,
2015 TRI data was the most current and
complete dataset. EPA has updated the
final risk evaluation document using 2018
TRI data in response to this comment and
the elapsed time since original scoping.
The 2018 TRI data is the most current and
complete data set as of January 2020. EPA
has updated the DMR dataset used in the
risk evaluation to reflect the latest available
data from 2018 as well.

85

Environmental Justice

•	EPA does not consider environmental justice despite Executive
Order 12898, which mandates "disproportionately high and
adverse human health or environmental effects of its programs,
policies, and activities on minority populations and low-income
populations" are identified and addressed. Executive Order 12898
does not contain exemptions for any type of agency "programs,
policies, and activities."

•	EPA has acknowledged that it must analyze risks to environmental
justice communities as part of its risk evaluation and include

TSCA § 6(b)(4) requires that EPA conduct a
risk evaluation to "determine whether a
chemical substance presents an unreasonable
risk of injury to health or the environment,
without consideration of cost or other non-risk
factors, including an unreasonable risk to a
potentially exposed or susceptible
subpopulation identified as relevant to the risk
evaluation by the Administrator, under the
conditions of use." TSCA § 3(12) states that

Page 13 of 212

-------
environmental justice communities in the development of the risk
evaluation. EPA must determine if exposure to 1,4-dioxane will
result in disproportionate risks to "minority populations and low-
income populations."

• Residents of low-income and communities of color may face
greater exposure to 1,4-dioxane, making EPA's failure to comply
with TSCA and EPA implementing regulations particularly
egregious from the perspective of environmental justice.

Page 14 of 212

"the term 'potentially exposed or susceptible
subpopulation' means a group of individuals
within the general population identified by the
Administrator who, due to either greater
susceptibility or greater exposure, may be at
greater risk than the general population of
adverse health effects from exposure to a
chemical substance or mixture, such as infants,
children, pregnant women, workers, or the
elderly." EPA believes that the statutory
directive to consider potentially exposed or
susceptible subpopulations (PESS) and the
statutory definition of PESS inherently include
environmental justice populations. Thus, EPA's
consideration of PESS in this risk evaluation
addresses the requirements of the Executive
Order.

EPA considers both exposure (Section 2.4) and
biological (Section 3.2.6.1) considerations in
evaluating PESS. As discussed in Section 4.4,
certain human subpopulations may be more
susceptible to exposure to 1,4-dioxane than
others. Some subpopulations may be more
biologically susceptible to the effects of 1,4-
dioxane due to genetic variability, pre-existing
health conditions, lifestage, pregnancy, or other
factors that alter metabolism or increase target
organ susceptibility. Other susceptibility factors
may include race/ethnicity and socioeconomic
status. The variability in human susceptibility
to 1,4-dioxane is reflected in the selection of
the uncertainty factor for human variability
included in the benchmark MOE. In addition,

-------




EPA accounts for exposures to PESS by using
the high-end exposure value when making its
unreasonable risk determinations.

EPA seeks to achieve the fair treatment and
meaningful involvement of any group,
including minority and/or low income
populations, in the development,
implementation, and enforcement of
environmental laws, regulations, and policies.
To this end, the Agency has already sought
input from specific populations and public
health experts in implementing TSCA and will
continue to do so. EPA will also consider
environmental justice populations in
accordance with the Executive Order as it
develops risk management actions based on
final TSCA section 6(b) risk evaluations.

Scope ( oininenls - Kxcluricri Conditions of I so

SACC,24,
47, 58

Byproducts

• The justifications for not including byproducts as a condition of use
were inadequate, and in breach of TSCA mandates. There is a
requirement to evaluate circumstances in which a chemical is
intended, known, or reasonably foreseen to be manufactured,
processed, distributed in commerce, used, or disposed § 2602(4).
Byproducts (or impurities or contaminants) are considered
conditions of use under TSCA. There is also a requirement to
consider all available information on exposures resulting from the
conditions of use of the chemical, without exception. See 15 U.S.C.
§ 2605(b)(4)(F)(i).

As explained in the scope document, 1,4-
dioxane may be found as a contaminant in
consumer products that are readily available for
public purchase. In the final risk evaluation,
eight consumer conditions of use are evaluated
based on the uses identified in EPA's 2015
TSCA Work Plan Chemical Problem
Formulation and Initial Assessment of 1,4-
Dioxane ( ). An additional
systematic review effort was undertaken for
consumer exposures to identify, screen, and
evaluate relevant data sources. These
conditions of use include use of 1,4-dioxane as
a surface cleaner, antifreeze, dish soap,

Page 15 of 212

-------




dishwasher detergent, laundry detergent, paint
and floor lacquer, textile dye, and spray
polyurethane foam (SPF). 1,4-Dioxane may be
found in these products at low levels (0.0009 to
0.02%) based on its presence as a byproduct of
other formulation ingredients {i.e., ethoxylated
chemicals). Inhalation exposures are estimated
for consumers and bystanders and dermal
exposures are estimated for users. Acute
exposures are presented for all consumer
conditions of use, while chronic exposures are
presented for the conditions of use that are
reasonably expected to involve daily use
intervals {i.e., surface cleaner, dish soap,
dishwasher detergent, and laundry detergent).
See Section 2.4.3 of the final risk evaluation.

51, 58

Use and Disposal of Fuel/Fuel Additives

• EPA unlawfully excluded the use and disposal of 1,4- dioxane as a
fuel or fuel additive because it determined that such uses had been
discontinued. Exclusion of these so-called "legacy uses" and
"legacy disposal" is unlawful under TSCA, because, even if these
uses have been discontinued, the ongoing disposal of these
products is still a circumstance under which the chemical is known
or reasonably foreseen to be disposed of. See 15 U.S.C. § 2602(4).

• The use of 1,4-dioxane in the past as a
racing fuel additive is not a "legacy" use.
As described in EPA's Risk Evaluation
Rule (82 FR 33726, 33729 (July 20,
2017)), a legacy use is an activity that does
not reflect ongoing or prospective
manufacturing, processing, or distribution
in commerce for that application. The
commenter appears to be describing
associated disposal or legacy disposal. The
example provided in the Risk Evaluation
Rule for associated disposal is the future
disposal of insulation that contains a
chemical substance, which may be present
in buildings after a chemical substance is
no longer being manufactured, processed,

Page 16 of 212

-------




or distributed for that use. In contrast, 1,4-
dioxane in racing fuel is no longer being
manufactured, processed, distributed in
commerce, used, or disposed of, to the best
of EPA's knowledge, which is based on
EPA's research and outreach and review of
reasonably available information; therefore,
this does not fall under the definition of
legacy use or associated disposal.
Specifically, EPA received no information
from any commenters or otherwise
indicating that racing fuel products in the
United States had been stockpiled or that
use or disposal was ongoing. Finally,
racing authorities have prohibited the use
of 1,4-dioxane in racing fuels, and EPA has
no information to suggest that it is, has
been, or would be used in fuels other than
racing fuels. Therefore, EPA does not
consider use or disposal of 1,4-dioxane in
racing fuel additive to be a condition of use
of 1,4-dioxane. Any disposal associated
with past use of 1,4-dioxane as a racing
fuel additive would be considered a "legacy
disposal" that has already occurred and
would also not be considered a condition of
use of 1,4-dioxane. See Safer Chemicals,
Healthy Families v. EPA, 943 F.3d 397
(9th Cir. 2019).

48, 52

Spray Polyurethane Foam

• EPA included spray foam in the risk evaluation; however, the
manufacturer claims that 1,4-dioxane is not used in the

• The commenter appears to be referring to
the Safety Data Sheet (SDS) and comments
submitted by Johns Manville (JM), a
manufacturer of spray polyurethane foam

Page 17 of 212

-------


manufacturing of the product and disagrees with the level of 1,4-
dioxane claimed to be in spray foam that was purportedly obtained
from the SDS. The manufacturer indicates that the product SDS
makes no mention of 1,4-dioxane and attempted to contact EPA
regarding the source of the reported value and did not get a
response.

(SPF), but this SDS was not used in the
risk evaluation because the product no
longer contains 1,4-dioxane. SDS from two
other companies were used. The JM
product is referenced, along with others, in
the Preliminary Information on
Manufacturing, Processing, Distribution,
Use, and Disposal: 1,4-Dioxane document.
The SDS, written in 2011 by JM, note
concentrations of-0.04% for 1,4-dioxane.
In subsequent years it appears JM may
have updated the product and/or SDS as the
current 2019 revision does not list 1,4-
dioxane as an ingredient (as the commenter
and manufacturer have pointed out).
Therefore, the risk evaluation does not
include the JM product because it does not
contain 1,4-dioxane. The risk evaluation
includes SPF products that do contain 1,4-
dioxane from other manufacturers as the
basis of potential worker exposures.

Scope ( oininenls - Kxcluricri Kxposurc Psilhwsivs

SACC,
19, 24,
42, 43,
44, 47,
50,51,
54, 55,
56, 58, 60

•	The justifications for excluding the general population, consumers,
and susceptible subpopulations, and for not including byproducts
as a condition of use, or evaluating risks to the environment were
inadequate, and in breach of TSCA mandates.

•	Regulatory Nexus: Provide additional scientific basis for how
general population, occupational and consumer exposures not
currently assessed under TSCA are effectively managed under
other regulatory authorities. TSCA authorizes EPA to consider its
other statutory authorities only in the risk management phase. See
15 U.S.C. § 2608.

•	The risk evaluation should be revised to include these populations,

EPA found that exposures to the general
population may occur from the conditions of
use due to releases to air, water or land. The
exposures to the general population via
drinking water, ambient air and sediment
pathways falls under the jurisdiction of other
environmental statutes administered by EPA,
i.e., CAA, SDWA, and RCRA. As explained in
more detail in section 1.4.2, EPA believes it is
both reasonable and prudent to tailor TSCA
risk evaluations when other EPA offices have

Page 18 of 212

-------
expertise and experience to address specific
environmental media, rather than attempt to
evaluate and regulate potential exposures and
risks from those media under TSCA. EPA has
therefore tailored the scope of the risk
evaluations for 1,4-dioxane using authorities in
TSCA sections 6(b) and 9(b)(1). EPA did not
evaluate hazards or exposures to the general
population via certain pathways in the risk
evaluation, and as such the unreasonable risk
determinations for relevant conditions of use do
not account for exposures to the general
population for certain pathways. However, the
final risk evaluation includes an evaluation of
general population exposures through
recreational activities {i.e., swimming) in
ambient water bodies. See Section 1.4.2 of the
final risk evaluation.

As explained in the scope document, 1,4-
dioxane may be found as a contaminant in
consumer products that are readily available for
public purchase. In the final risk evaluation,
eight consumer conditions of use are evaluated
based on the uses identified in EPA's 2015
TSCA Work Plan Chemical Problem
Formulation and Initial Assessment of 1,4-
Dioxane ( ). An additional
systematic review effort was undertaken for
consumer exposures to identify, screen, and
evaluate relevant data sources. These
conditions of use include use of 1,4-dioxane as
a surface cleaner, antifreeze, dish soap,	

Page 19 of 212

or EPA needs to address each relevant TSCA mandate and provide
credible justification with empirical evidence that demonstrates the
exposures in question are irrelevant and, therefore, not covered
under TSCA law.

• The risk evaluation should be revised to include the air and

drinking water exposure pathways, and also evaluation of ingestion
of contaminated food products.

-------




dishwasher detergent, laundry detergent, paint
and floor lacquer, textile dye, and spray
polyurethane foam (SPF). 1,4-Dioxane may be
found in these products at low levels (0.0009 to
0.02%) based on its presence as a byproduct of
other formulation ingredients {i.e., ethoxylated
chemicals). See Section 2.4.3 of the final risk
evaluation.

TSCA section 3(2) defines "chemical
substance" to exclude any food or food additive
as the terms are defined in section 201 of the
Federal Food, Drug, and Cosmetic Act, when
manufactured, processed, or distributed in
commerce for use as a food or food additive.
Therefore, EPA believes that the ingestion of
contaminated food products falls under the
jurisdiction of FDA.

Kxposiirc Assn 111 pi ions

SACC,
21, 22,
46, 58

EPA's consideration of compliance with OSHA's worker protection
standards is not clear. Specific examples and citations are required.
EPA evaluated worker risk assuming PPE is used to mitigate exposure
and risk. To comply with TSCA's worker protection mandate, EPA
must evaluate 1,4-dioxane without assuming any PPE use.

The OSHA regulations at 29 CFR 1910.132
require employers to assess a workplace to
determine if hazards are present or likely to be
present which necessitate the use of personal
protective equipment (PPE). If the employer
determines hazards are present or likely to be
present, the employer must select the types of
PPE that will protect against the identified
hazards, require employees to use that PPE,
communicate the selection decisions to each
affected employee, and select PPE that properly
fits each affected employee. OSHA has
established a permissible exposure limit (PEL)
of 100 ppm (8-hour TWA) for 1,4-dioxane.

Page 20 of 212

-------
However, as noted on OSHA's website, "OSHA
recognizes that many of its permissible
exposure limits (PELs) are outdated and
inadequate for ensuring protection of worker
health. Most of OSHA's PELs were issued
shortly after adoption of the Occupational
Safety and Health (OSH) Act in 1970, and have
not been updated since that timeOSHA
provides an annotated list of PELs on its
website, including alternate exposure levels.
For 1,4-dioxane, the alternates provided are the
California OSHA PEL of 0.28 ppm and the
ACGIH TLV of 20 ppm.
(https://www.osha.gov/dsg/annotated-
pels/tablez-l.html) EPA's approach for
evaluating risk to workers and ONUs is to use
the reasonably available information and
professional judgement to construct exposure
scenarios that reflect the workplace practices
involved in the conditions of use of the
chemicals . When appropriate, in the risk
evaluation, EPA will use exposure scenarios
both with and without engineering controls
and/or PPE that may be applicable to particular
worker tasks on a case-specific basis for a
given chemical. Thus, while EPA has evaluated
worker risk with and without PPE, as a matter
of policy, EPA does not believe it should
assume that workers are unprotected by PPE
where such PPE might be necessary to meet
federal regulations, unless it has evidence that
workers are unprotected.

Page 21 of 212

-------




For the purposes of determining whether or not
a condition of use presents unreasonable risks,
EPA incorporates assumptions regarding PPE
use based on information and judgement
underlying the exposure scenarios. These
assumptions are described in the unreasonable
risk determination for each condition of use, in
section 5.2 of the risk evaluation. Additionally,
in consideration of the uncertainties and
variabilities in PPE usage, EPA uses the high-
end exposure value when making its
unreasonable risk determination in order to
address those uncertainties. EPA has also
outlined its PPE assumptions in section 5.1 of
the risk evaluation. Further, in the final risk
evaluation for 1,4-dioxane, EPA has
determined that most conditions of use pose an
unreasonable risk to workers even with the
assumed PPE.

Kisk ( hiirnclcriznlion (ommcnls

5 5

• There is a requirement to consider aggregate exposure See TSCA
Section 6(b)(4)(F).

TSCA section (•>(!">)(4)( 1 )(ii) directs N\\ to

"describe whether aggregate or sentinel
exposures to a chemical substance under the
conditions of use were considered, and the
basis for that consideration" in risk evaluations.
EPA defines aggregate exposures as the
combined exposures to an individual from a
single chemical substance across multiple
routes {i.e., dermal, inhalation, or oral) and
across multiple pathways {i.e., exposure from
different sources). 40 CFR 702.33. EPA defines
sentinel exposures as the exposure from a
single chemical substance that represents the

Page 22 of 212

-------




plausible upper bound of exposure relative to
all other exposures within a broad category of
similar or related exposures. 40 CFR 702.33.
EPA considered the reasonably available
information and used the best available science
to determine whether to consider aggregate or
sentinel exposures for a particular chemical.
EPA has determined that using the high-end
risk estimate for inhalation and dermal risks
separately as the basis for the unreasonable risk
determination is a best available science
approach. There is low confidence in the result
of aggregating the dermal and inhalation
exposures and risks for this chemical if EPA
uses an additive approach, due to the
uncertainty in the data that could be reliably
modeled into the aggregate exposure such as
would occur with a PBPK model. Using an
additive approach to aggregate exposure and
risk in this case would result in an overestimate
of risk. Given all the limitations that exist with
the data, EPA's approach is the best available
approach.

Kisk Dolormin;ilion Comments

48

• It is unclear how EPA's risk characterization supports its risk
determination. The risk characterization summary discussion
requires a description of how new information impacts the risk
characterization.

• See the Executive Summary, updated Risk
Characterization (Section 4), and updated
Risk Determination (Section 5) for more
clarification on how these sections support
each other and how new information is
incorporated.

SACC, 47

• Provide a clear scientific rationale for the determination of
"reasonable" risk for conditions of use that require personal
protective equipment for such determination to be appropriate.

While EPA believes that discussions of the
rationale for the determination of unreasonable
risk is outside the scope of the SACC, EPA is

Page 23 of 212

-------
There is a requirement to make risk determinations based on
hazard and exposure, not consideration of "nonrisk" factors. See
15 U.S.C. § 2605(b)(4)(A), (F)(iii).

Page 24 of 212

committed to providing the public with
sufficient information on the basis for that
determination. TSCA requires EPA to
determine whether chemicals in the
marketplace present unreasonable risks to
health or the environment. While the law does
not specifically define this term unreasonable
risk, during the risk evaluation process EPA
weighs a variety of factors including the effects
of the chemical on human health or the
environment, populations exposed (including
any sensitive subpopulations), the severity of
the hazard, and uncertainties. This approach is
outlined in EPA's 2017 Procedures for
Chemical Risk Evaluation Under the Amended
Toxic Substances Control Act rule ("Risk
Evaluation Rule") preamble on how risk
evaluations will be conducted. [82 FR 33726, at
33735 (July 20, 2017)] Each draft risk
evaluation details those factors and describes
for the public which conditions of use were
preliminarily identified to have unreasonable
risk for a chemical. For 1,4-dioxane, these
factors included a range of workplace
exposures.

EPA's approach for developing exposure
assessments for workers is to use the
reasonably available information and expert
judgment. When appropriate, in the risk
evaluation, EPA will use exposure scenarios
both with and without engineering controls
and/or personal protective equipment (PPE)

-------
that may be applicable to particular worker
tasks on a case-specific basis for a given
chemical. Again, while EPA has evaluated
worker risk with and without PPE, as a matter
of policy, EPA does not believe it should
assume that workers are unprotected by PPE
where such PPE might be necessary to meet
federal regulations, unless it has evidence that
workers are unprotected. For the purposes of
determining whether a condition of use presents
unreasonable risks, EPA incorporates
assumptions regarding PPE use based on
information and judgment underlying the
exposure scenarios. These assumptions are
described in the unreasonable risk
determination for each condition of use, in
Section 5.2. For example, in the case of 1,4-
dioxane, which is manufactured, processed, and
used in industrial settings, where there are
typically strong industrial hygiene programs
that include training and oversight, EPA
believes that it is reasonable to assume a
protection factor (PF) of 10 or 20 for dermal
protection (gloves) and assigned protection
factor (APF) of 25 or 50 for inhalation
protection (respirators). For 1,4-dioxane, each
condition of use includes a characterization of
risks at the central tendency and high-end
exposures and also when PPE was considered
at these exposure levels. EPA presented each of
these risk estimates to the public in the draft
risk evaluation, and refined them with
additional information for the final risk

Page 25 of 212

-------




evaluation. For the purposes of determining
whether or not a condition of use presents an
unreasonable risk, EPA incorporates
assumptions regarding PPE use based on
information and judgement underlying the
exposure scenarios. These assumptions are
described in the unreasonable risk
determination for each condition of use, in
section 5.2. Additionally, in consideration of
the uncertainties and variabilities in PPE usage,
EPA uses the high-end exposure value when
making its unreasonable risk determination in
order to address those uncertainties. In the final
risk evaluation, EPA has determined that most
of the conditions of use present unreasonable
risks to workers even with the assumed PPE.

SACC, 22

Incorporate the tabular format for the risk determination as done in
Section 6. The final risk determination sections should be clarified.
EPA should cite the relevant supporting scientific information
(section/page/table numbers from the draft risk evaluation) for each
decision made under the risk determination section.

• While EPA is unable to add such extensive
citations, the formatting and clarity changes
to the unreasonable risk determination
section from draft to final should provide
the reader with an understanding of which
risk characterization tables are relevant for
each condition of use. Similarly, the new
summary table in the risk characterization
chapter (section 4) provides a unified
crosswalk between major pieces of
information throughout the risk evaluation
for each condition of use.

ChirilY ( oniincnls

22, 40,
48, 58

•	The tiered assessment approach used for exposure assessment
should be described more clearly.

•	A clear statement indicating whether or not EPA had access to the
full study reports for all of the studies cited is needed.

• EPA added language to Executive

Summary of the risk evaluation describing
its approach for exposure assessment,
which is also discussed in section 2.4.

Page 26 of 212

-------
EPA's assumptions in modeling exposures to 1,4-dioxane would
benefit from additional documentation.

Page 27 of 212

EPA's approach for developing exposure
assessments for workers and ONUs is to
use the reasonably available information
and professional judgment. When
appropriate in the risk evaluation, EPA has
used exposure scenarios both with and
without engineering controls and/or PPE
that may be applicable to particular worker
tasks on a case-specific basis for a given
chemical. While EPA has evaluated worker
risk with and without PPE, as a matter of
policy, EPA does not believe it should
assume that workers are unprotected by
PPE where such PPE might be necessary to
meet federal regulations, unless it has
evidence that workers are unprotected. For
the purposes of determining whether or not
a condition of use presents unreasonable
risks, EPA incorporates assumptions
regarding PPE use based on information
and judgment underlying the exposure
scenarios. These assumptions are described
in the unreasonable risk determination for
each condition of use, in section 5.2.
Additionally, in consideration of the
uncertainties and variabilities in PPE usage,
EPA uses the high-end exposure value
when making its unreasonable risk
determination in order to address those
uncertainties. EPA has also outlined its
PPE assumptions in section 5.1. Further, in
the final risk evaluation for 1,4-dioxane,
EPA has determined that most conditions

-------
of use pose an unreasonable risk to workers
even with the assumed PPE.

• EPA provides detailed explanations and
sample calculations for modeled exposures
in Appendix G. This appendix provides a
rational and basis for the parameters used,
assumptions made, and a narrative of the
various models. All parameters, equations,
and methods are cited where applicable for
further analysis and review.	

Page 28 of 212

-------
2. Systematic Review

Charge Question 2.1: Please comment on the approaches and/or methods used to support and inform the gathering, screening,
evaluation, and integration of data/information used in the Draft Risk Evaluation for 1,4-Dioxane.

Charge Question 2.2: Please also comment on the clarity of the information as presented related to systematic review and suggest
improvements as warranted.

Summary of Comments lor Specific
Issues Related lo Charge Question 2

KPA/OPPT Response

SACC,
48,56,
58

General Comments

• The application of systematic review criteria is
inconsistent. There is a reliance on some sources that do
not go through the systematic review process, while
other sources are excluded "on the basis of the
systematic review process." The inclusion or exclusion
of sources appears in some cases to be based on the
decision about whether or not to apply the systematic
review process, and these decisions are not fully
explained or justified.

• All cited data sources should be made publicly available.

• While some sources did not go through the initial
inclusion/exclusion process, all data used in the risk
evaluation went through data evaluation criteria and
received a data quality rating.

• To the extent possible, EPA makes the studies it relies
on publicly accessible via the EPA HERO database.
Each citation in the risk evaluation is hyperlinked to its
HERO database entry. Most journal article entries in
HERO have a link to a DOI (Digital Object Identifier).
This link will direct you to a journal or publisher
website. If the article is free to the public, or you have a
subscription to the journal, you can download the PDF.
If not, you will usually be offered an option to purchase
the individual article. Copyright law prohibits EPA
from distributing copyrighted material.

IX IIA

)ossiers

• It is not clear why studies cited in ECHA dossiers would
bypass the data screening step and move directly to the
data evaluation step.

• It is not transparent whether full studies were obtained,
or whether EPA has relied on industry-prepared

• Studies previously identified by authoritative sources
such as ECHA were automatically identified as relevant
for consideration. Additional studies were identified
through systematic review searches. All studies then
went through data quality evaluation to determine

Page 29 of 212

-------


summaries without access to the full studies. EPA
should request full studies from the submitters.

• A clear distinction must be made between industry data
that have not been evaluated, industry data that have
been evaluated by ECHA or other government
authorities in the EU, and information that ECHA has
itself developed or provided.

whether they are acceptable for inclusion in the risk
evaluation, regardless of how they were initially
identified. EPA only used complete study reports as key
and supporting studies and did not rely on industry-
prepared summaries. On their own, the robust
summaries available through ECHA do not provide
sufficient information to receive an "acceptable" rating
in EPA's data quality evaluation process. While EPA
relies on previous assessments to help identify
potentially useful sources of data, it does not rely on
previous assessments to determine the quality of those
sources. EPA evaluated all of the key and supporting
studies it relied on in this risk evaluation using its own
data quality evaluation process.

Qusililv Kvnliisilion

22, 24,
48, 56,
58, 59

•	There is no empirical basis for the current scoring
method to exclude research based on a single reporting,
or methodological limitation.

•	Submitters disagree with OPPT's systematic review
methodology wherein if a single metric is assigned a
score of Unacceptable, the entire study is dismissed.

•	Professional judgement was used to upgrade or
downgrade overall study scores for animal toxicity data
in four instances.

•	Efforts need to be taken to calibrate the reviews of
different reviewers, as some inconsistencies in data
quality evaluation both within and across chemicals
seems apparent.

o Staff doing the data quality evaluations must
have appropriate subject matter expertise and be
trained on general data quality review methods.

•	Appendix A of the Application of Systematic Review in
TSCA Risk Evaluations explains the basis for
EPA/OPPT's development of a numerical scoring
system to inform the characterization of the
data/information sources during the data integration
phase. The intent is to provide transparency and
consistency to the evaluation process along with
creating evaluation strategies that meet the TSCA
science standards for various data/information streams.

•	EPA/OPPT's quality evaluation method was developed
following identification and review of various
published qualitative and quantitative scoring systems
to inform our own fit-for-purpose tool. The
development process involved reviewing various
evaluation tools/frameworks (e.g., NTP's Office of

Page 30 of 212

-------
Efforts being made to do internal quality checks
on the data quality evaluations for individual
studies and risk evaluations must be disclosed.

Page 31 of 212

Health Assessment and Translation (OHAT) Risk of
Bias tool, Criteria for Reporting and Evaluating
Ecotoxicity Data (CRED), etc.; see Table 1 and
Appendix A of the Application of Systematic Review in
TSCA Risk Evaluations and references therein), as well
as soliciting input from scientists based on their expert
knowledge about evaluating various data/information
sources for risk assessment purposes. While there are
many published systematic review tools available for
human health and environmental health hazard
assessment, no systematic review tools were identified
that encompass either exposure assessment (e.g.,
general population exposures, occupational exposures
and industrial releases) or fate and transport assessment.
The data quality evaluation results published with each
risk evaluation provides the lists of references
EPA/OPPT evaluated for the first 10 TSCA risk
evaluations.

In order to ascertain the quality of the available data,
EPA/OPPT used a numerical scoring system to assign a
qualitative rating. The goal of this approach was to add
consistency and transparency to the evaluation process.
Scores were used for the purpose of assigning the
confidence level rating of High, Medium, Low, or
Unacceptable, and informed the characterization of
data/information sources during the data integration
phase. The data quality evaluation results for the first
ten TSCA Risk Evaluations are posted on chemical
specific websites. In all evaluation strategies,
professional judgment was employed to determine the
adequacy or appropriateness of the qualitative rating

-------
assigned by the numerical scoring system.

•	The TSCA evaluation strategies consider
methodological design and implementation and
reporting within the existing domains and metrics.

Since it is difficult to have high confidence in data
where the underlying methods are unreported or poorly
reported, EPA assesses reporting and methodological
quality simultaneously. However, EPA recognizes the
challenge of discerning between a deficit in reporting
and a problem in the underlying methodological quality
of the data/information source. Developing a reporting
checklist, guidance document or a separate reporting
quality domain may be a future solution for
consideration in optimizing the evaluation strategies.
EPA also designed evaluation criteria that consider risk
of bias and Bradford Hill aspects when assessing the
quality of animal toxicity and epidemiological studies.
Refer to Appendices F, G and H of the Application of
Systematic Review in TSCA Risk Evaluations for more
information.

•	Relevant data sources are evaluated for data quality
following title/abstract and full-text screenings for the
first 10 and next 20 TSCA risk evaluations, after a pilot
period to calibrate criteria and revise as needed.
Generally, each study evaluation is conducted by at
least two reviewers, with a process for comparing and
resolving differences. This helps ensure quality
assurance. However, based on assessment needs, the
assessment team should make decisions about how
many reviewers are needed. While more than one
reviewer is ideal, there may be times when one

Page 32 of 212

-------
reviewer is acceptable, such as when the assessment
needs to be conducted under a rapid timeframe and the
outcome being reviewed is unlikely to be a driver for
the assessment. These quality assurance methods are
the same as used by EPA's IRIS Program. Other EPA
Offices (such as Office of Research and Development
and the Office of Science Coordination and Policy)
partnered with OPPT in developing innovations in
searching and screening for the next 20 chemical
evaluations (see response to Q5) and continues to
support OPPT in scoping and SR efforts.

• The data evaluation is conducted in a tool (e.g., Excel,
Access, DistillerSR) that tracks and records the
evaluation for each data/information source including
reviewer's comments. The evaluation results for each
study evaluated under TSCA were released publicly
with each draft risk evaluation to validate the evaluation
tools and explore potential differences in professional
judgment that may arise from multiple reviewers (both
internal and external stakeholders). This documentation
approach also increased transparency of professional
judgment calls to stakeholders and the public for the
first 10 TSCA risk evaluations. EPA/OPPT plans to
use these evaluation strategies, including pre-
determined criteria, documented in EPA's Application
of Systematic Review in TSCA Risk Evaluations
document, for the next 20 TSCA risk evaluations.
However, refinements to the evaluation strategies are
likely to occur. EPA already made changes to the
physical chemical properties, environmental hazard,
and epidemiological criteria since the Application of

Page 33 of 212

-------




Systematic Review in TSCA Risk Evaluations
document was published. These changes were due to
validation and improvement efforts to ensure that the
most relevant studies were included in the TSCA risk
evaluations, and the most up-to-date data quality
evaluation criteria are used for the next 20 TSCA risk
evaluations

SACC,
56, 58,
59

Potential Bias

• EPA relies on voluntary submissions for much of its
exposure data. There is concern that this could lead to a
collection of biased data or submission of data that is
"cherry picked." Additional steps should be taken to
ensure that the information received is accurate and
complete. A process should be in place for vetting
statements and assertions made by entities with a
financial interest in the outcome of the risk assessment.

• EPA evaluated data submitted using the data evaluation
criteria. However, in the future, EPA will put all data
submitted to the Agency through a screening process
and then an evaluation process that utilizes the same
criteria as data identified from literature searches.

24, 56,
58

Epidemiological Data Quality Criteria

• Under the new criteria, epidemiological studies can no
longer receive high scores for all study metrics, making
it difficult for epidemiological studies to be scored
overall as high quality.

o Some metrics (#18-22) no longer allow a score of

unacceptable,
o No explanations or empirical support was
provided for the revisions to the systematic
review data quality criteria for epidemiological
studies.

•	EPA/OPPT's quality evaluation method was developed
following identification and review of various
published qualitative and quantitative scoring systems
to inform our own fit-for-purpose tool. The
development process involved reviewing various
evaluation tools/frameworks (e.g., OHAT Risk of Bias
tool, CRED, etc.; see Appendix A of the Application of
Systematic Review in TSCA Risk Evaluations
document and references therein), as well as soliciting
input from scientists based on their expert knowledge
about evaluating various data/information sources
specifically for risk assessment purposes.

•	The epidemiologic criteria were later revised to more
stringently distinguish between High, Medium and Low
studies. After additional piloting of the criteria, EPA

Page 34 of 212

-------
found that the initial iteration of the epi data quality
criteria (as published in the Application of Systematic
Review in TSCA Risk Evaluations) was inadvertently
skewing quality scores toward the tail ends of the
scoring spectrum (High and Unacceptable). In order to
have the criteria represent a more accurate depiction of
the quality levels in the epi literature, the criteria were
revised using 2 methods.

•	The first method was to make the unacceptable metrics
less stringent. This was accomplished by either
rewording the metrics to allow for more professional
judgement in the interpretation of the unacceptable
criterion, or in some cases, completely removing the
unacceptable bin from metrics that EPA determined
were not influential enough to completely disqualify a
study from consideration (mostly metrics in the
Analysis and Biomonitoring domain). EPA found that
these criteria changes greatly reduced the type one error
in the Unacceptable scoring. No acceptable studies
were inaccurately classified as Unacceptable.

•	The second method was to reduce the number of studies
that received an overall High rating. The majority of
overall scores in EPA's initial evaluations during
piloting tended to be High. Therefore, EPA strived to
revise the criteria to provide more gradation in the
scoring to more accurately and objectively distinguish
studies of the highest quality from medium and low
quality studies. To do this, EPA removed the High
criterion from some metrics, particularly in
dichotomous metrics (High/Low or High/Unacceptable)
that were primarily being binned as High by reviewers

Page 35 of 212

-------




across the majority of the studies. These dichotomous
metrics were contributing to the overall quality scores
being skewed towards High. To address this, EPA
shifted some of the dichotomous metrics such that the
highest metric score possible (for all studies) is a
Medium. The change led to the dichotomous metrics
having less significant impact to the numerical scoring
and the overall quality rating for each study.

• With the aforementioned changes to the criteria, EPA
observed fewer studies with Unacceptable ratings and
more studies shifting from High to Medium, with only
the highest quality studies receiving a High overall
rating. Out of the -200 relevant epidemiologic studies
and cohorts evaluated for data quality for the first 10
TSCA chemicals, the majority (-80%) still scored as
High or Medium. The remaining -20% of studies
scored Low or Unacceptable. EPA is confident that no
studies of acceptable quality were inappropriately
assigned as Unacceptable. EPA is also confident that
the revised criteria bins the quality levels of these epi
studies more appropriately than the previous iteration.
Additional refinements to the epidemiologic data
evaluation criteria are likely to occur as EPA's
validation and process improvement efforts continue.

InU'grsilion

24, 48,
58

•	The approach to evaluation of quality is clearly
addressed in the supplemental documents for the risk
evaluation, but the approaches for evaluation of
consistency, relevance, coherence, and biological
plausibility are not as clearly documented.

•	While EPA provides the overall study quality ratings for

• EPA will work with the National Academy of Sciences,
Engineering, and Medicine (NASEM) TSCA
Committee to consider revisions to the data quality
evaluation criteria and options regarding integrating
evidence within and across evidence streams (human,
animal, mechanistic data). EPA proposes to use a more

Page 36 of 212

-------
the pertinent studies for each endpoint, it is not clear
how final conclusions were reached for all hazard
endpoints.

The failure to provide a pre-established protocol for
evidence integration and instead relying on a "weight-of-
the-scientific evidence narrative" does not align with
best practices shared by leading systematic review
methods.

A specific weight-of-evidence methodology should be
presented.

EPA should more clearly and transparently present
biologically robust, weight of evidence assessments
where data integration is required, such as the
Organization for Economic Cooperation and
Development (OECD) Adverse Outcome Pathway
(AOP) methodology or the mode of action (MOA)
approach initially championed by the World Health
Organization (WHO)/International Program on
Chemical Safety (IPCS).	

Page 37 of 212

structured framework for evidence integration for the
next set of chemicals evaluated under TSCA.

-------
3. Environmental Fate, Exposure & Effects

Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if any, should be considered?

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

Dismissal of Available Kiivironmenlal Kxposure Data

SACC

Tile Committee recommended inclusion of all reasonably
reliable data for aqueous 1,4-Dioxane concentrations, 1,4-
Dioxane concentrations in sediment, and aquatic toxicity
results with aggregate weighting factors related to the quality
of each study. This approach will reward studies of the highest
quality, while not ignoring studies that may be outliers or that
were performed in an era when the current record keeping rules
were not established. For example, extant data describing 1,4-
Dioxane in surface water could be used rather than modeled
surface water concentrations.

In EPA's 2018 Problem Formulation, ambient surface
water monitoring data from STORET and NWIS were
noted to range from 0.568 to 100 |ig/L. EPA also
conducted screening-level aquatic exposure modeling
during problem formulation that informed the decision
not to further analyze the pathway during risk
evaluation. Predicted levels of 1,4-dioxane in surface
water from this screen were as high as 11,500 |ig/L that
was linked to releases from facilities. EPA included
additional sources, as identified in SACC and public
comments, in the risk evaluation to better characterize
surface water levels of 1,4-dioxane that aquatic species
could be exposed to. While none of these reported levels
exceeded levels predicted from EFAST modeling, they
are now referenced for a more robust characterization of
1,4-dioxane in surface waters in the risk evaluation.

24, 58

For environmental exposure data, EPA did not use the best
available science as directed by TSCA. Available empirical
data were not always considered, and in some instances, there
was a reliance on outdated or incorrect data. Exclusion of
reasonably available information is contrary to TSCA policy.

EPA conducted qualitative and quantitative analyses in
the problem formulation stage that informed the level of
environmental analysis in the risk evaluation. Based on
this effort during problem formulation, environmental
exposure pathways for ecological receptors were not
further analyzed during risk evaluation. However, EPA
incorporated the analysis in the final risk evaluation with
updates to correct TRI release information, incorporate
indirect dischargers, and apply an updated acute COC.
Reasonably available surface water discharge,

Page 38 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





environmental fate, and ecological toxicity data were
utilized during problem formulation to support the
decision to conduct no further analysis for
environmental exposure pathways.

42, 43,
60, 61,
62

Submitters provided multiple data monitoring points indicating
1,4-dioxane levels in rivers, drinking water, ground water, and
surface water that are orders of magnitude above EPA advisory
levels; these empirical data were not considered in the Risk
Evaluation. This includes 105 site investigation reports
measuring 1,4-dioxane's presence in the environment.

•	Samples were collected from: wastewater treatment plant
discharges, Cape Fear River, Haw River, intake water for
Pittsboro's drinking water treatment plant, ground water
and landfill leachate in Minnesota, and treated wastewater
effluent.

•	Available data reporting 1,4-dioxane levels in sludge from
a manufacturing facility in Fayetteville, NC report high
levels, establishing exposure from biosolids can be
significant. These data were not considered.

Drinking water exposures to the general population via
surface and/or groundwater sources were not within the
scope of this evaluation (See Section 1.4.2 of the final
risk evaluation). However, these data in surface water
were used in the final risk evaluation to estimate
incidental oral and dermal exposure to the general
population from recreational activities (i.e., swimming)
in ambient water.

Regarding the aquatic exposure assessment, while the
referenced surface water data may indicate levels in the
environment above EPA and/or state advisory levels,
they were not greater than the predicted (modeled)
estimates used during problem formulation to determine
that no further analysis on the aquatic exposure
assessment to ecological receptors was warranted during
risk evaluation. Predicted levels of 1,4-dioxane in
surface water, as reported in the final problem
formulation and the final risk evaluation, are as high as
11,500 |ig/L for acute scenarios, whereas the highest
level of 1,4-dioxane reported in the submitted surface
water sources is 1,405 |ig/L.

• EPA included additional surface water monitoring
data sources as identified in SACC and public
comments, to better characterize surface water levels

Page 39 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

of 1,4-dioxane that aquatic species could be exposed
to. These data cover measured levels of 1,4-dioxane
in the Cape Fear Watershed; Cape Fear, Deep, and
Haw Rivers (including near the Pittsboro drinking
water intake), and Minnesota surface waters. While
none of these reported levels exceeded levels
predicted from EFAST modeling, they are now
referenced for a more robust characterization of 1,4-
dioxane in surface waters. The referenced Minnesota
landfill leachate data were not utilized based on the
scope of the evaluation (See Section 1.4.2).

• The value of 20.4 ppm measured in North Carolina
likely represents an extreme case, and was measured
in sludge, not dewatered and processed biosolids.
Even so, it is only somewhat higher than the chronic
aquatic COC of 14.5 ppm (an aqueous concentration,
so comparison with the sludge value would require
correction for moisture content). So, while a
hypothetical direct exposure to this sludge may be
associated with ecological risk, the risk for the
indirect pathway of runoff from land-applied
biosolids is extremely low.

24, 58

• The 2015 data used did not include indirect discharges to
water. The justification for excluding this data was
insufficient and did not meet the TSCA mandate to use "the
best available science."

• 2016-2018 TRI data, which report higher water release
values than the 2015 document, were not included in the
risk evaluation. This resulted in an underestimation of the

• EPA's first-tier aquatic exposure modeling effort
initially did not include indirect discharges or
transfers off-site for waste treatment. In response to
public comment, EPA has augmented this first-tier
analysis to include indirect dischargers (i.e., facilities
reporting off-site transfers to POTW for treatment)
for the years analyzed (2014-2015).

Page 40 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

impact from water releases.

• The modeling done during problem formulation to
conduct the analysis of environmental exposure for
aquatic species used the two most recent complete
years of TRI reporting at the time of problem
formulation - 2014 and 2015. EPA's risk evaluation
was informed by the decision not to further analyze
environmental pathways set forth in its problem
formulation.

24, 58

• Air, land, and disposal or other release data from the 2015
Toxic Releases Inventory (TRI) data were not considered
under the false assumption that these would be regulated by
other environmental statutes.

• Air emission values reported for 1,4-dioxane through the
National Emissions Inventory (NEI), which are much
higher than those reported under the TRI, were not cited or
evaluated.

• According to the 2018 Toxics Release Inventory data, a
number of manufacturing facilities emit 1,4- dioxane to the
air and release it to the water. These releases were not
included in the risk evaluation.

• Section 1.4.2 in the risk evaluation provides details
as to why certain pathways were not included in the
risk evaluation. However, because there is no
nationally recommended Ambient Water Quality
Criteria under the CWA, EPA included exposures to
the general population via ambient surface water.
EPA evaluated hazards and exposures to the general
population from ambient surface water for the
conditions of use in the risk evaluation. The final
risk evaluation includes 1,4-dioxane water releases
based on 2018 TRI and DMR reporting. These
releases were used to model ambient water
concentrations and estimate incidental oral and
dermal exposure to the general population from
recreational activities (i.e., swimming).

• Air releases were not included in the scope of the
risk evaluation, as stated in Section 1.4.2.

• In its revision, EPA included a release assessment
describing releases to water. EPA has augmented its
original first-tier aquatic exposure assessment to
include additional facilities releasing and reporting to

Page 41 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

TRI for the years analyzed during problem
formulation (2014-2015). The revised first-tier
aquatic exposure assessment includes the direct and
indirect discharging facilities.

EPA did not consider data from the Third Unregulated
Contaminant Monitoring Rule ("UCMR3") that includes
comprehensive monitoring data.

Data from UCMR3, which provides nationally
representative data on the occurrence of contaminants in
drinking water, were not utilized in the aquatic exposure
assessment due to a focus on ambient surface water
levels since general population drinking water exposures
were not included in the scope of the risk evaluation (see
Section 1.4.2).

SACC

On page 46: Clarity is increased by listing the "conservative"
assumptions in estimating values for surface water in a bulleted
form. Unregulated Contaminant Monitoring Rule (UCMR-3)
data can serve this purpose (Adamson etal., 2017). Similarly,
surface water values (secondary and tertiary wastewater
values) from the State of California can also be used for
Measured Environmental Concentrations (MEC) (Anderson et
al., 2018). Estimated concentrations can be placed in context
by using UCMR data. Additionally, Simonich etal., (2013)
determined that 38 of 40 wastewater treatment plant (WWTP)
discharges contained detectable 1,4-Dioxane amounts, but at
lower concentrations than modeled. This suggests that sorption
or volatilization from WWTPs may not have been adequately
assessed to protect workers and the broader population from
1,4-dioxane inhalation or exposure to biosolids.

Data from UCMR3, which provides nationally
representative data on the occurrence of contaminants in
drinking water, were not utilized in the aquatic exposure
assessment due to a focus on ambient surface water
levels since general population exposures via drinking
water were not included in the scope of the risk
evaluation (see Section 1.4.2).

In its revision of the draft risk evaluation, EPA included
additional sources to characterize surface water levels of
1,4-dioxane that aquatic species could be exposed to.
While none of these reported levels exceeded levels
predicted from EFAST modeling, they are now
referenced for a more robust characterization of 1,4-
dioxane in surface waters.

Removal by volatilization may vary depending on
WWTP design, but exposures via air are out of scope for

Page 42 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

this risk evaluation because 1,4-dioxane is regulated
under the Clean Air Act (See Section 1.4.2). EPA
expects that sorption of 1,4-dioxane to sludge will be
negligible in every WWTP due to its organic carbon-
water partition coefficient (log Koc = 0.4), but 1,4-
dioxane is expected to be present in the biosolids-
associated water at concentrations similar to the bulk
water in the sludge settling tank. Direct human
exposures to biosolids are not expected, but exposures
resulting from land-applied biosolids are expected to be
via air from volatilized 1,4-dioxane or via drinking water
as a result of surface runoff from biosolids-treated land.
However, exposures via air and drinking water are out of
scope for this risk evaluation because 1,4-dioxane is
regulated under the Clean Air Act and Safe Drinking
Water Act (see Section 1.4.2).

272 on-topic studies identified as relevant to analyzing aquatic
exposures were not evaluated.

As described in the caption to Figure 2-6 in the draft risk
evaluation, EPA determined during problem formulation
that environmental exposure pathways for ecological
receptors were within scope but would not be further
analyzed based on quantitative and qualitative analyses
covering ecological pathways (U.S. EPA, 2018c). These
analyses were made ahead of the data screening stage for
these data sources, and therefore, the environmental
exposure references were excluded, as they did not meet
the risk evaluation PECO statement.

EPA cannot disregard data (or consider it to be zero) where the
analytical sample values have extremely high method detection
limits (MDLs).

EPA thanks the commenter for pointing out this
consideration of the STORET data cited in the draft risk
evaluation. While EPA did consider the referenced range

Page 43 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

• The justification for discarding MDLs that were nearly
double the chronic aquatic COC was invalid.

from STORET for years 2007 through 2017 in its draft
risk evaluation, the RQs presented in Table 5-2 are
dependent on results of the screening-level modeling
analysis. Additionally, though some of the sampling
MDLs were higher than the chronic COC, EPA did not
use unreported/unknown levels below such MDLs as the
basis for an RQ or unreasonable risk determination. In
response, EPA has augmented its discussion of
uncertainty in Section 4.3.2 with a discussion of this
point.

SACC

Using measured surface water concentrations is particularly
important. The surface water data range of 0.5-100 [j,g/L
appears to be erroneously low. Information published in 2016
by Sun, Loez, and Knappe, found 543 [j,g/L in one sample from
the Cape Fear River, North Carolina and over 1,400 ug/L in
WWTP effluent. The 543 [j,g/L concentration in surface water
was determined using an EPA method (Sun etal., 2016). These
surface water measurements exceed the Agency exposure
estimate for surface water by a factor of 4.4-5.4. The
Committee recommended including these data in the estimates
of chronic exposure and factoring these into the final risk.
These values are useful in estimating the distribution function
used in estimating higher percentile concentrations.

The measured surface water concentrations cited from
STORET would not include individual published data
sources, such as those noted by the SACC panel.
Therefore, they may reflect lower levels than those
reported elsewhere based on differences in sampling
methods and sampling location, e.g., proximity to
sources of 1,4-dioxane in surface water. EPA did not
conduct a comprehensive review of surface water
monitoring data for 1,4-dioxane on the basis that the
screening-level modeling conducted during problem
formulation supported performing no further analysis.
The 543 |ig/L cited from Sun et al., (2016) does not
exceed either the chronic or acute concentrations of
concern (COCs) derived in the ecological hazard
assessment for 1,4-dioxane, which are 57,500 and
14,500 |ig/L, respectively. 543 |ig/L is also less than the
high-end exposure concentrations estimated in EFAST,
which included a maximum estimate of 11,500 |ig/L for
a 10-day, acute release scenario.

Page 44 of 212

Summary of Comments for Specific
Issues Uclalcd lo Charge Question 3

KIW/OPPT Response

I nsupporlcd Assumptions Uclalcd lo Knvironnienlal Kxposurc

24, 55,
58

The data to support the following claims were not provided in

the risk evaluation:

• "Recent monitoring data on ambient surface water levels
indicate relatively low levels." (p. 213)

• EPA acknowledges "[T]here are relatively fewer data
available on 1,4-dioxane levels in surface water," (p. 28),
indicating a data gap that EPA apparently will do nothing
to address.

• "Limited sediment monitoring data for 1,4-dioxane that are
available, suggest that 1,4-dioxane is present in sediments."
(pp. 131,211).

• In Section 3.3.1, EPA states "National-scale
monitoring data from EPA's STOrage and
RETreival (STORET) and National Water
Information System (NWIS) for the past ten years,
shows that 1,4-dioxane is detected in surface water.
The data points show a detection rate of
approximately 6% for this media, with detections
ranging from 0.568 to 100 (J,g/L."

• EPA acknowledges that it did not include additional
references to ambient surface water levels based on
its approach using a screening-level aquatic exposure
assessment during problem formulation supportive
of not doing further analysis during risk evaluation.
In its revision of the draft risk evaluation, EPA
included additional sources to characterize surface
water levels of 1,4-dioxane that aquatic species could
be exposed to. While none of these reported levels
exceeded levels predicted from EFAST modeling,
they are now referenced for a more robust
characterization of 1,4-dioxane in surface waters.

Page 45 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

• As described in Section 3.1 and Section 5.1.1, EPA's
assessment of the fate of 1,4-dioxane in sediment
was based on a qualitative consideration of its
physical-chemical properties. Based on its log Kow
and water solubility, 1,4-dioxane in sediment will be
present primarily in the sediment porewater as
opposed to sorbed to the solids.

SACC

In the Evaluation (page 21, and also Table 2-8 in the Problem
Formulation document (U.S. EPA 2018), EPA states it "did not
identify any exceedances of benchmarks to aquatic vertebrates,
aquatic invertebrates, and aquatic plants from exposures to 1,4-
Dioxane in surface waters." Missing from this list is discussion
of toxicity to benthic organisms. Adverse effects were assessed
for only one aquatic invertebrate species (Evaluation page 80).
The absence of benthic organism data represents a serious data
gap, as does the absence of multiple chronic toxicity studies for
any species or guild (Kaviraj et al., 2004, Bernot el al., 2005,
Saha et al., 2006, Dobbins et al., 2009, Guo et al., 2012, Chen
et al., 2018, Liu et al., 2018, Yang et al., 2018, Ibrahim and
Sayed 2019, Kang et al., 2019). Sediment organisms have quite
different sensitivities to many toxicants than are surface
invertebrates, and bivalves are often much more sensitive
(Kaviraj et al., 2004, Liu et al., 2018, Dobbins et al., 2009).
The assumption of similar toxicity to other species has
questionable merit. Please note that the needed data can be
obtained within the time frame of risk assessment finalization.

EPA recognizes that benthic/sediment-dwelling
organisms are highly sensitive to various xenobiotics
and the hazard could be very different from those that
dominate in the water column. However, after
examining the physical/chemical properties of 1,4-
dioxane, EPA concludes that the toxicity will be low to
sediment-dwelling organisms that are exposed to pore
water and sediment-dwelling species that were
characterized for hazard in the surface water can be used
as a surrogate species. EPA stands by the rationale for
using the toxicity profile for Daphnia magna and
Gammarus pseudolimnaeus to read-across for 1,4-
dioxane's effects to sediment dwelling organisms. It has
been well documented that D. magna has been used to
study the effects of hazardous chemicals to pore water
and sediment contaminants such as metals and organic
compounds (Giesy et al., 1998; Othoudt etal., 1991;
Ristola etal., 1995; Coen and Janssen, 1998; Suedel and
Rodgers 1996; CPA, 2004; Parkake^a/., 2010). Other
surface water species such as Brachionus calyciflorus
and Thamnocephalusplaxyurus (Heida and Oost, 1996)
have also been used to determine the effects of

Page 46 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

hazardous chemicals in pore water. Although these
procedures and protocols have been updated, the results
from these studies have been used for establishing
benchmark dose levels for sediment toxicity.

24, 50,
55, 58

It is unclear how EPA can state that there is a lack of
unreasonable risks to the environment (pp. 21, 156) when
significant data gaps were identified, including ecotoxicity data
for soil or sediment dwelling organisms, plants, terrestrial
species, and avian species, and a lack of aquatic chronic
toxicity data except for fish.

• EPA should use its information authorities through TSCA
to address identified data gaps.

EPA derived environmental concern levels based on
hazard values from highly acceptable studies and
reported exposure levels in the environment. EPA stands
by the analysis that was conducted to determine the
hazard and risk of 1,4-dioxane to the environment. After
a complete analysis of the hazard data of 1,4-dioxane,
EPA is confident that the risks of this chemical are low
to the aquatic and terrestrial organisms. These
conclusions are supported by other countries that have
investigated the hazard of 1,4-dioxane.

Page 47 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

study the effects of hazardous chemicals to pore water
and sediment contaminants such as metals and organic
compounds (Giesy et al., 1998; Othoudt etal., 1991;
Ristola etal., 1995; Coen and Janssen, 1998; Suedel and
Rodgers 1996; CPA, 2004; Parkake^a/., 2010). Other
surface water species such as Brachionus calyciflorus
and Thamnocephalusplaxyurus (Heida and Oost, 1996)
have also been used to determine the effects of
hazardous chemicals in pore water. Although these
procedures and protocols have been updated, the results
from these studies have been used for establishing
benchmark dose levels for sediment toxicity.

24, 58

The water release value reported in the risk evaluation is not
the total water release value of 56,935 lbs as reported in TRI
2015. EPA removed discharges to sewage treatment plants
from the total water release; an explanation for this decision
should be provided.

EPA's intention was not to exclude indirect discharges
or transfers to wastewater treatment sites (e.g., POTWs)
from its first-tier screening level aquatic exposure
assessment as described in the problem formulation. In
response to this comment, EPA compared the TRI
release information (direct and indirect discharges) used
during problem formulation to TRI release information
extracted from EPA's TRI Explorer website and
identified some inconsistencies. EPA found that some of
the site-specific off-site waste transfers to POTWs for
treatment were of a greater magnitude than the site-
specific direct discharges modeled for years 2014-2015
during problem formulation. Additionally, some of the
facilities originally included as direct dischargers were
also identified as errors. Therefore, EPA has corrected
its initial modeling assessment of TRI direct dischargers
and added sites reporting off-site waste transfers to

Page 48 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

POTWs for treatment. The modeling is still done on a
site-specific basis. The updates and corrections are
shown in Appendix E of the final risk evaluation.

It was assumed that no exposures were expected to occur
during distribution because chemicals are packaged in closed
system containers. Data/documentation to support this
assumption must be provided.

• Are drums, bottles, and pails that may be opened still
considered "closed" systems?

The chemical packagers and end-users incorporated
closed systems into their chemical management
solutions.

Closed system containers, unless otherwise intentionally
opened, are sealed during transport, distribution and
handling to prevent accidental releases. Thus, closed
system containers under normal operation will not result
in exposure to 1,4-dioxane.

EPA states that 1,4-dioxane "is not likely to accumulate in
wastewater biosolids..." (p. 45), and that "exposures to surface
water from biosolids are estimated to be low" (p. 131, 212);
support for this assumption should be provided.

• To accurately assess impacts to the environment from land-
applied biosolids, a total accounting of the 1,4-dioxane in
the biosolids should be developed.

• The assumption that land-applied biosolids are only
generated through wastewater treatment plants (WWTP) is
incorrect.

o Available data reporting 1,4-dioxane levels in

sludge from a manufacturing facility in Fayetteville,
NC report high levels, establishing exposure from
biosolids can be significant. These data were not
considered.

• It cannot be assumed that, just because 1,4-dioxane does
not partition strongly to organic material, that there is no

• EPA's assessment of the fate of 1,4-dioxane in land-
applied biosolids was based on a qualitative
consideration of its physical-chemical properties,
combined with a comparison to the scenario of direct
surface water discharge. Given its high water
solubility and low Henry's law constant and log Koc,
1,4-dioxane in wastewater treatment influent will be
associated primarily with the water phase and will
not volatilize or adsorb to solids. 1,4-Dioxane in
biosolids will be present in the porewater of the
material. The production of biosolids from waste
streams involves dewatering processes, which will
fractionate 1,4-dioxane and result in lower volume
concentrations of the compound in biosolids
compared to those found in influent and effluent
waters. Less than 2% of 1,4-dioxane in wastewater
treatment plant influent is expected to be present in

Page 49 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

pathway for exposure via sediment or land-applied
biosolids. The empirical fact that 1,4-dioxane is present in
these media must be considered.

biosolids. Thus, while leaching from land-applied
biosolids does represent a plausible pathway for
transport of 1,4-dioxane, fractionation in wastewater
treatment and environmental dilution mean that the
masses introduced via this pathway will be far lower
than those associated with direct discharge to surface
water. As assessed elsewhere in the risk evaluation,
surface water concentrations are predicted to be well
below the chronic COC for aquatic organisms. The
value of 20.4 ppm (in solids) measured in North
Carolina likely represents an extreme case, and was
measured in sludge, not fully processed biosolids.
Even so, it is only somewhat higher than the chronic
aquatic COC of 14.5 ppm (an aqueous concentration,
so comparison with the sludge value would require
correction for moisture content). So, while a
hypothetical direct exposure to this sludge may be
associated with ecological risk, the risk for the
indirect pathway of runoff from land-applied
biosolids is extremely low.

• Based on its log Koc (0.4) and water solubility (>800
mg/L), 1,4-dioxane does not partition to soil and
sediment. In a Level III fugacity model assuming
100% of emissions to soil, it is estimated that 7% of
1,4-dioxane will be in soil at equilibrium, 93% will
be in air, and <0.1% will be in water or sediment.
Thus, based on physical/chemical and fate
properties, 1,4-dioxane may be expected to be
present in soil and sediment, but exposure from soil

Page 50 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

and sediment is likely to be negligible compared to
exposure via air.

43, 50

The risk evaluation states the physical-chemical properties of
1,4-dioxane were used to determine that sediment, soil and
biosolids were not relevant pathways. It is not clear that any
investigations were done to identify empirical monitoring data.
If concentrations were detected in these media, then the
pathways should have been evaluated.

• Industrial biosolids delivered to a North Carolina

composting plant contained 1,4-dioxane at 20,000 parts per
billion—a high level considering that EPA's health
advisory for the chemical is 35 parts per billion in drinking
water.

EPA conducted qualitative and quantitative analyses in
the problem formulation stage that informed the level of
environmental analysis in the draft risk evaluation.

Based on this effort during problem formulation,
environmental exposure pathways for ecological
receptors were not further analyzed during risk
evaluation.

• The production of biosolids from waste streams
involves dewatering processes, which will
fractionate 1,4-dioxane and result in lower volume
concentrations of the compound in biosolids
compared to those found in influent and effluent
waters. Less than 2% of 1,4-dioxane in wastewater
treatment plant influent is expected to be present in
biosolids. Thus, while leaching from land-applied
biosolids does represent a plausible pathway for
transport of 1,4-dioxane, fractionation in wastewater
treatment and environmental dilution mean that the
masses introduced via this pathway will be far lower
than those associated with direct discharge to surface
water. The value of 20.4 ppm (in solids) measured in
North Carolina likely represents an extreme case,
and was measured in sludge, not fully processed
biosolids.

24, 58

EPA states that measured and estimated levels of 1,4-dioxane
in the environment are sufficiently below the acute and chronic

This conclusion was based on the comparison of
estimated (modeled) surface water levels of 1,4-dioxane

Page 51 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

aquatic concentrations of concern (COCs) without providing
analysis to support this conclusion.

to acute and chronic COCs, included in Section 4.1 of
the draft risk evaluation.

EPA disagrees with the commenter's conclusion that the
analysis for deriving the concern levels in the aquatic
environment are not supported. EPA has made revisions
in the environmental hazard section of the risk
assessment that clarifies the methods used to derive the
acute and chronic concentrations of concern. Also,
during problem formation, EPA conducted a preliminary
assessment regarding the hazard and risk of 1,4-dioxane
to aquatic receptors. EPA identified the following
sources of environmental hazard data for 1,4-dioxane:
(Health Canada, 2010; ECJRC, 2002; OECD, 1999;
NICNAS, 1998); and the European Chemicals Agency
(ECHA) Database. Studies published since 2003 were
identified in the literature search for 1,4-dioxane and
were reviewed as described in Application of Systematic
Review in TSCA Risk Evaluations (U.S. EPA, 2018a)
and Strategy for Assessing Data Quality in TSCA Risk
Evaluations (U.S. EPA, 2018b). These studies have been
summarized in section 2.4 Hazards (Effects) of the risk
evaluation. The data show that 1,4-dioxane's toxicity is
low to aquatic organisms. Based on the
physical/chemical properties of 1,4-dioxane, the
chemical is expected to migrate through soil and the
final fate will be to groundwater. These conclusions
have been in the environmental fate, hazard and risk
characterization sections of the risk evaluation.

Page 52 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

24, 58

EPA's modeling of surface water concentrations includes
assumptions that are not necessarily conservative, despite
EPA's claims to the contrary.

• For example, EPA points to the surface water modeling
assumption that "[wastewater treatment removal is
assumed to be 0% for this exercise" (p. 29); yet its own
modeling of wastewater treatment removal efficiency using
EPISuite STP module indicates removal rates will be very
low, on the order of 2% (p. 24). Far from being a
conservative assumption, this use of 0% is a reasonable
conclusion based on the available data.

• Despite a promised "full table of results, see Appendix E"
(p. 29), that table provides only EPA's conclusions and
none of its analysis.

• EPA states that "wastewater treatment removal is
assumed to be 0% for all direct discharges, as
reported direct loadings/releases are assumed to
account for any pre-release treatment" in Appendix
E of the final risk evaluation. This is not a claim of
conservatism, but an explanation of modeling inputs.

• Table E-3 in Appendix E includes the values needed
to cross-check results using the publicly available
model used. Additionally, the supplemental file
published along with the draft risk evaluation
contains detailed facility information for additional
context. All sites modeled are shown, along with
release inputs and key results (concentration based
on low-flow 7Q10 conditions and predicted days of
exceedance of the chronic COC for non-acute release
scenarios [i.e., those with 20 days or more of
release]). Because the predicted surface water
concentrations based on the low-flow (7Q10)
conditions are typically used by OPPT to assess
ecological risk, this was the only predicted
concentration in Tables E-3 through E-5. EPA
understands that the table does not show certain
elements that would be more comprehensive.
Therefore, in the final risk evaluation, EPA has
included an EFAST output report for a modeled
scenario as an example of the source of the values in
the result table. EPA has also included in its final
risk evaluation a column showing the conversion
from reported annual loads in lbs/year to kg/year,

Page 53 of 212

Summary of C omments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

which were used in EFAST modeling. These updates
are included in the supplemental file: Aquatic
Exposure Screen Facility Information.

r.nvironnienlal Assessment Methodology

SACC

The rationale for using modeled surface water data rather than
measured data is unclear. The relative contribution of
"estimated" and "predicted" modeling values seem to describe
similar processes and are simply different tools to model
concentrations that are unknown or assumed to be unknown. It
is also unclear how ambient data were used to confirm model
estimates. Given the databases available, there would be
greater certainty to use Measured Environmental
Concentrations (MECs) rather than Predicted Environmental
Concentrations (PECs) for risk assessments.

EPA did not conduct a comprehensive review of surface
water monitoring data for 1,4-dioxane on the basis that
the screening-level modeling conducted during problem
formulation supported no further analysis during risk
evaluation. In this way, the modeling done during
problem formulation was treated as a first-tier of a tiered
assessment approach.

In Section 3.3.1, EPA states "National-scale monitoring
data from EPA's STOrage and RETreival (STORET)
and National Water Information System (NWIS) for the
past ten years, shows that 1,4-dioxane is detected in
surface water. The data points show a detection rate of
approximately 6% for this media, with detections
ranging from 0.568 to 100 (J,g/L." EPA acknowledges
that it did not include additional references to ambient
surface water levels based on its approach using a
screening-level aquatic exposure assessment during
problem formulation supportive of not doing further
analysis during risk evaluation. In the final risk
evaluation, EPA included additional sources to better
characterize surface water levels of 1,4-dioxane that
aquatic species could be exposed to. While none of these
reported levels exceeded levels predicted from EFAST

Page 54 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

modeling, they are now referenced for a more robust
characterization of 1,4-dioxane in surface waters.

SACC

While Monte Carlo analyses was used to incorporate variability
and examine uncertainty in estimates obtained from models for
inhalation exposure estimates. Monte Carlo methods were not
similarly used in the examination of estimates of environmental
exposures. No reasons were given for this decision. Regardless
of whether Monte Carlo methods were used, uncertainty in the
estimates presented for environmental exposures needs to be
addressed in the Evaluation.

• Monte Carlo analyses should be included in

environmental estimates, and this will require a robust
data set.

In Section 4.3.2 of the risk evaluation, EPA
characterizes uncertainty surrounding the prediction of
surface water concentrations using EFAST. While EPA
did not conduct a quantitative uncertainty analysis for
the model itself, the risk evaluation discusses uncertainty
of the key inputs and incorporates variability by
modeling multiple release day scenarios for each facility
modeled, resulting in a range of surface water estimates.
The key modeling input that drives surface water
concentration estimates is the release volume
(kg/site/day). Many of the facility discharges are
reported in TRI or DMR as a single annual loading
estimate; there may not be an available range to consider
modeling to capture site-specific variability in release
volume. Therefore, from facilities reporting these data
on a site-specific basis. The variability stems from the
uncertainty surrounding possible annual release days,
which EPA did consider by modeling 1, 20, and 250
days of release for direct dischargers. EPA's approach is
conservative, as the risk characterization relied on the
highest modeled surface water concentrations (i.e., those
associated with the lowest release day scenario for a
given discharger).

• EPA develops Monte Carlo analyses in

environmental estimates when it is appropriate.
For 1,4-dioxane, first-tier analyses indicated that

Page 55 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

a more comprehensive analysis (i.e., Monte
Carlo) was not warranted.

SACC

The Agency erroneously avoided using the aqueous
concentration upper bound (11,500 (J,g/L: Table E-3) for the
chronic aquatic environmental exposure assessment. The
11,500 [j,g/L value is not an acute value it was a 10-day average
(Table E-3; footnote b) and it approaches the chronic toxicity
threshold (14,500 [j,g/L effect). Unlike all other facilities for
which releases were modeled, the DAK facility was not
considered for the single day release. Considering only the 10-
day release scenario decreased acute surface concentration
estimates by factor of 10. Thus, neither a worst case nor high
percentile estimate is presented in this assessment. Using the
upper bound is an appropriate choice given the modeled nature
of this exposure estimate. Using 10 x 11,500 [j,g/L (to account
for the unmodeled single day release rather than the modeled
10-day release) would produce an acute RQ of 0.46.

As the SACC panel pointed out, a footnote explains that
a 10-day scenario was utilized for this site based on
engineering assumptions related to the lowest number of
operating days for a site falling within this standard
industrial category. However, based on EPA's standard
procedures for new chemicals, a 10-day release scenario
is still considered acute in nature and would still be
compared against the acute COC. Only releases of 20
days per year or more are compared against the chronic
COC.

Scope o

' r.nvironnienlal Kxposurc Assessment

SACC

Exposure scenarios that include consumers should be included
in the 1,4-dioxane hazard determination. The presence of 1,4-
Dioxane in plastic, other commercially available products,
surface water, drinking water, groundwater, and in sediments is
well documented and the risks to human health are as yet
unassessed by the Agency. The American Grocers and the
Cleaning Products Institute (both trade associations) agree.

Regarding consumers, as explained in the scope
document for 1,4-dioxane, 1,4-dioxane may be found as
a contaminant in consumer products and/or commercial
products that are readily available for public purchase.
However, it is present as a result of byproduct formation
during manufacture of ethoxylated chemicals that are
subsequently formulated into products. EPA did not
evaluate exposures to consumers and bystanders from
byproduct or contaminant exposure in this risk
evaluation. In the final risk evaluation, eight consumer
conditions of use are evaluated based on the uses

Page 56 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

identified in EPA's 2015 TSCA Work Plan Chemical
Problem Formulation and Initial Assessment of 1,4-
Dioxane ( 015). An additional systematic
review effort was undertaken for consumer exposures to
identify, screen, and evaluate relevant data sources.
These conditions of use include use of 1,4-dioxane as a
surface cleaner, antifreeze, dish soap, dishwasher
detergent, laundry detergent, paint and floor lacquer,
textile dye, and spray polyurethane foam (SPF). 1,4-
Dioxane may be found in these products at low levels
(0.0009 to 0.02%) based on its presence as a byproduct
of other formulation ingredients (i.e., ethoxylated
chemicals). Inhalation exposures are estimated for
consumers and bystanders and dermal exposures are
estimated for users. Acute exposures are presented for
all consumer conditions of use, while chronic exposures
are presented for the conditions of use that are
reasonably expected to involve daily use intervals (i.e.,
surface cleaner, dish soap, dishwasher detergent, and
laundry detergent). See Section 2.4.3 of the final risk
evaluation.

Please refer to section 1.4.2 in the risk evaluation which
provides details as to why certain pathways of exposure
to the general population were not included in the scope
of the risk evaluation.

SACC

General human population and biota exposure must be assessed
by the Agency for inhalation, ingestion, and dermal routes of
exposure within the defined time limit for a TSCA assessment.

Environmental exposures via surface water, sediment,
and biosolids pathways were quantitatively or
qualitatively assessed during problem formulation.
Please refer to section 1.4.2 in the risk evaluation which

Page 57 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

This appropriately broader population should include different
sensitive or highly exposed sub-populations.

The Agency should consider human exposures from: lawn
watering, public pools, and dust abatement at construction sites
or on roads. The concentrations of 1,4-Dioxane in sediments
should be explicitly summarized or tabulated, not noted as
present and dismissed (Problem Formulation: page 41).

provides details as to why certain pathways of exposure
to the general population were not included in the scope
of the risk evaluation. These excluded pathways include
human exposures from the activities such as lawn care,
etc. which would be covered by SDWA. However,
because there is no nationally recommended Ambient
Water Quality Criteria under the CWA, EPA included
exposures to the general population via ambient surface
water. EPA evaluated hazards and exposures to the
general population from ambient surface water for the
conditions of use in the risk evaluation. The final risk
evaluation includes 1,4-dioxane water releases based on
2018 TRI and DMR reporting. These releases were used
to model ambient water concentrations and estimate
incidental oral and dermal exposure to the general
population from recreational activities (i.e., swimming).

SACC

Exposure assessment through groundwater and other
environmental pathways must be evaluated. Data on these
pathways should be generated if unavailable. Groundwater is
regulated by the Clean Water Act only if it is used for
municipal purposes. Omission of groundwater in the exposure
assessment means that risks to consumers of groundwater are
unknown. Data are available to define the numbers of
individuals consuming groundwater from private wells for
drinking water and/or irrigation of crops. These data can be
used in conjunction with subsurface injection site location
information to provide estimates of the numbers of potentially
exposed individuals. In many areas, groundwater is directly

Please refer to section 1.4.2 in the risk evaluation which
provides details as to why certain pathways of exposure
to the general population were not included in the scope
of the risk evaluation. Exposures to the general
population via drinking water, which includes finished
surface and ground water are covered under SDWA. In
addition, 1,4-dioxane is a hazardous waste injected into
Class 1 underground injection hazardous wells under the
jurisdiction of RCRA. Section 1.4.2 in the risk
evaluation provides details as to why these pathways
were not included in the risk evaluation.

Page 58 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

recharged from surface waters (example: Edwards Aquifer in
Texas) further increasing the numbers of potentially exposed.
On page 46, the Agency should determine to what extent
groundwater is contaminated by the million pounds of 1,4-
Dioxane injected into subsurface zones over the past several
years (Table E-l: Class 1 Underground injection column). This
human exposure must be considered, given the current use and
the fact that millions of U.S. citizens and residents consume
and otherwise utilize groundwater from unregulated wells.

SACC

The frequency of detects in drinking water is less important
than the number of persons who are exposed to those
concentrations (Problem Formulation: page 43). A population
exposure estimate should be provided in the assessment. To
accomplish this, data would be needed for large, medium and
small water management facilities as well as differing water
types, soft, moderately hard, near brackish, effluent dominated,
so forth. This should have been pointed out as a data need in
the problem formulation process. The omission of exposure
through drinking water leaves the 1,4-Dioxane Evaluation
incomplete.

Please refer to section 1.4.2 in the risk evaluation which
provides details as to why this pathway was not included
in the risk evaluation.

19, 24,
43, 50,
58

EPA's reasoning for the exclusion of some environmental
exposure pathways is based on the assumption that they will be
adequately assessed and managed through other statutes or
regulations (including the Resource Conservation and
Recovery Act [RCRA], the Clean Air Act [CAA], the Clean
Water Act [CWA], the Safe Drinking Water Act [SDWA], and
various state programs). However:

• EPA did not show or establish that these regulations
eliminate any unreasonable risk.

EPA found that exposures to the general population may
occur from the conditions of use due to releases to air,
water or land. The exposures to the general population
via drinking water, ambient air and land pathways falls
under the jurisdiction of other environmental statutes
administered by EPA, i.e., CAA, SDWA, CERCLA, and
RCRA. As explained in more detail in section 1.4.2,
EPA believes it is both reasonable and prudent to tailor
TSCA risk evaluations when other EPA offices have

Page 59 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

• No statute that addresses groundwater as a source of
exposure was identified, and groundwater or groundwater
discharge to surface water are not identified as exposure
pathways in the risk evaluation.

• Disposal is not limited to RCRA Subtitle C landfills.

• Many of these statutes and regulations vary by state and/or
are not adequately enforced (e.g., RCRA).

• EPA recognized that 1,4-dioxane has been detected in
landfill leachate but fails to identify adequate management
under state or federal laws.

expertise and experience to address specific
environmental media, rather than attempt to evaluate and
regulate potential exposures and risks from those media
under TSCA. EPA has therefore tailored the scope of the
risk evaluations for 1,4-dioxane using authorities in
TSCA sections 6(b) and 9(b)(1). EPA did not evaluate
hazards or exposures to the general population via
drinking water, ambient air, or sediment pathways in the
risk evaluation, and as such the unreasonable risk
determinations for relevant conditions of use do not
account for exposures to the general population.
However, because there is no nationally recommended
Ambient Water Quality Criteria under the CWA, EPA
included exposures to the general population via ambient
surface water. EPA evaluated hazards and exposures to
the general population from ambient surface water for
the conditions of use in the risk evaluation. The final risk
evaluation includes 1,4-dioxane water releases based on
2018 TRI and DMR reporting. These releases were used
to model ambient water concentrations and estimate
incidental oral and dermal exposure to the general
population from recreational activities (i.e., swimming).

To justify its exclusion of exposures from air emission
pathways, in the problem formulation EPA merely provides a
list of technology-based standards for certain source categories.
EPA provides no analysis whatsoever as to: the extent to which
the standards cover the full range of stationary sources of this
chemical; the extent and magnitude of releases of the chemical
allowed under each of the standards; the duration, intensity,
frequency, and number of exposures resulting from those
allowable emissions (as required under TSCA section
6(b)(4)(F)(iv)); or any other factors that would be necessary to
analyze and determine the extent and nature of potential risk
allowed under the standards. EPA has not acknowledged, let
alone analyzed, the overall risks to the general population or to
vulnerable subpopulations due to the combination of exposures
arising from the various sources for which standards exist, not
to mention additional emission sources not subject to any
standard. EPA has made no attempt to reconcile any such risk
with that allowed under TSCA. In the absence of such
analyses, there is no basis whatsoever for EPA to assert that air

Page 60 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response



releases of this chemical have been adequately assessed or that
any risks have been effectively managed under TSCA's
standards.



43

Aquatic species were evaluated by using estimated discharges
of 1,4-dioxane to surface water from wastewater facilities; this
does not include releases from spills or leaching from
contaminated sites into groundwater and subsequent transport
to surface water. It also does not include data available for
wastewater releases.

Spills and leaks generally are not included within the
scope of a TSCA risk evaluation because in general they
are not considered to be circumstances under which a
chemical substance is intended, known or reasonably
foreseen to be manufactured, processed, distributed,
used, or disposed of. To the extent there may be
potential exposure from spills and leaks, EPA is also
declining to evaluate environmental exposure pathways
addressed by other EPA-administered statutes and
associated regulatory programs.

First, EPA does not identify 1,4-dioxane spills or leaks
as "conditions of use." EPA does not consider 1,4-
dioxane spills or leaks to constitute circumstances under
which 1,4-dioxane is manufactured, processed,
distributed, used, or disposed of, within TSCA's
definition of "conditions of use." Congress specifically
listed discrete, routine chemical lifecycle stages within
the statutory definition of "conditions of use" and EPA
does not believe it is reasonable to interpret
"circumstances" under which 1,4-dioxane is
manufactured, processed, distributed, used, or disposed
of to include uncommon and unconfined spills or leaks
for purposes of the statutory definition. Further, EPA
does not generally consider spills and leaks to constitute

Page 61 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





"disposal" of a chemical for purposes of identifying a
COU in the conduct of a risk evaluation.

In addition, even if spills or leaks of 1,4-dioxane could
be considered part of the listed lifecycle stages of 1,4-
dioxane, EPA has "determined" that spills and leaks are
not circumstances under which 1,4-dioxane is intended,
known or reasonably foreseen to be manufactured,
processed, distributed, used, or disposed of, as provided
by TSCA's definition of "conditions of use," and EPA is
therefore exercising its discretionary authority under
TSCA Section 3(4) to exclude 1,4-dioxane spills and
leaks from the scope of the 1,4-dioxane risk evaluation.
The exercise of that authority is informed by EPA's
experience in developing scoping documents and risk
evaluations, and on various TSCA provisions indicating
the intent for EPA to have some discretion on how best
to address the demands associated with implementation
of the full TSCA risk evaluation process. Specifically,
since the publication of the Risk Evaluation Rule, EPA
has gained experience by conducting ten risk evaluations
and designating forty chemical substances as low- and
high-priority substances. These processes have required
EPA to determine whether the case-specific facts and the
reasonably available information justify identifying a
particular activity as a "condition of use." With the
experience EPA has gained, it is better situated to
discern circumstances that are appropriately considered
to be outside the bounds of "circumstances... under

Page 62 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





which a chemical substance is intended, known, or
reasonably foreseen to be manufactured, processed,
distributed in commerce, used, or disposed of' and to
thereby meaningfully limit circumstances subject to
evaluation. Because of the expansive and potentially
boundless impacts that could result from including spills
and leaks as part of the risk evaluation, (e.g., due to the
unpredictable and irregular scenarios that would need to
be accounted for, including variability in volume,
frequency, and geographic location of spills and leaks;
potential application across multiple exposure routes and
pathways affecting myriad ecological and human
receptors; and far-reaching analyses that would be
needed to support assessments that account for
uncertainties but are based on best available science),
which could make the conduct of the risk evaluation
untenable within the applicable deadlines, spills and
leaks are determined not to be circumstances under
which 1,4-dioxane is intended, known or reasonably
foreseen to be manufactured, processed, distributed,
used, or disposed of, as provided by TSCA's definition
of "conditions of use."

Exercising the discretion to not identify spills and leaks
of 1,4-dioxane as a COU is consistent with the discretion
Congress provided in a variety of provisions to manage
the challenges presented in implementing TSCA risk
evaluation. See e.g., TSCA Sections 3(4), 3(12),
6(b)(4)(D), 6(b)(4)(F). In particular, TSCA Section

Page 63 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





6(b)(4)(F)(iv) instructs EPA to factor into TSCA risk
evaluations "the likely duration, intensity, frequency,
and number of exposures under the conditions of
use...suggesting that activities for which duration,
intensity, frequency, and number of exposures cannot be
accurately predicted or calculated based on reasonably
available information, including spills and leaks, were
not intended to be the focus of TSCA risk evaluations.
And, as noted in the preamble to the Risk Evaluation
Rule, EPA believes that Congress intended there to be
some reasonable limitation on TSCA risk evaluations,
expressly indicated by the direction in TSCA Section
2(c) to "carry out [TSCA] in a reasonable and prudent
manner."

For these reasons, EPA is exercising this discretion to
not consider spills and leaks of 1,4-dioxane to be COUs.

Second, even if 1,4-dioxane spills or leaks could be
identified as exposures from a COU in some cases, these
are generally not forms of exposure that EPA expects to
consider in risk evaluation. TSCA Section 6(b)(4)(D)
requires EPA, in developing the scope of a risk
evaluation, to identify the hazards, exposures, conditions
of use, and potentially exposed or susceptible
subpopulations the Agency "expects to consider" in a
risk evaluation. This language suggests that EPA is not
required to consider all conditions of use, hazards, or
exposure pathways in risk evaluations. EPA has chosen

Page 64 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





to tailor the scope of the risk evaluation to exclude spills
and leaks_in order to focus analytical efforts on those
exposures that present the greatest potential for risk.

In the problem formulation documents for many of the
first 10 chemicals undergoing risk evaluation, EPA
applied the same authority and rationale to certain
exposure pathways, explaining that "EPA is planning to
exercise its discretion under TSCA 6(b)(4)(D) to focus
its analytical efforts on exposures that are likely to
present the greatest concern and consequently merit a
risk evaluation under TSCA...." This approach is
informed by the legislative history of the amended
TSCA, which supports the Agency's exercise of
discretion to focus the risk evaluation on areas that raise
the greatest potential for risk. See June 7, 2016 Cong.
Rec., S3519-S3520.

In addition to TSCA Section 6(b)(4)(D), the Agency also
has discretionary authority under the first sentence of
TSCA Section 9(b)(1) to "coordinate actions taken under
[TSCA] with actions taken under other Federal laws
administered in whole or in part by the Administrator."
TSCA Section 9(b)(1) provides EPA authority to
coordinate actions with other EPA offices, including
coordination on tailoring the scope of TSCA risk
evaluations to focus on areas of greatest concern rather
than exposure pathways addressed by other EPA-
administered statutes and regulatory programs, which

Page 65 of 212

-------
Charge Question 3.1: Please comment on the data, approaches and/or methods used to characterize exposure to aquatic
receptors in surface water. What other additional information, if anv. should be considered?

#

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response





does not involve a risk determination or public interest
finding under TSCA Section 9(b)(2). EPA has already
tailored the scope of this risk evaluation using such
discretionary authorities with respect to exposure
pathways covered under the jurisdiction of other EPA-
administered statutes and associated regulatory programs
(see section 1.4.3).

Following coordination with EPA's Office of Land and
Emergency Management (OLEM), EPA has found that
exposures of 1,4-dioxane from spills and leaks fall under
the jurisdiction of RCRA. See 40 CFR 261.33(d)
(defining in part a hazardous waste as "any residue or
contaminated soil, water or other debris resulting from
the cleanup of a spill into or on any land or water of any
commercial chemical product or manufacturing
chemical intermediate having the generic name listed
[40 CFR 261.33(e) or (f)], or any residue or
contaminated soil, water or other debris resulting from
the cleanup of a spill, into or on any land or water, of
any off-specification chemical product and
manufacturing chemical intermediate which, if it met
specifications, would have the generic name listed in [40
CFR 261.33(e) or (f)]"); 40 CFR 261.33(f) (listing 1,4-
dioxane as hazardous waste no. U108). As a result, EPA
believes it is both reasonable and prudent to tailor the
TSCA risk evaluation for 1,4-dioxane by declining to
evaluate potential exposures from spills and leaks, rather
than attempt to evaluate and regulate potential exposures

Page 66 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

from spills and leaks under TSCA. See Section 2.4.1 in
the final risk evaluation for further explanation on a
COU basis.

24, 55

The risk evaluation does not evaluate 1,4-dioxane's presence as
a byproduct or contaminant; therefore, exposure as a result of
its presence in personal care and cleaning products was not
evaluated. Down-the-drain discharges and industrial discharges
to sewage treatment plants can contribute to groundwater and
surface water contamination.

As explained in the scope document, 1,4-dioxane may be
found as a contaminant in consumer products that are
readily available for public purchase. In the final risk
evaluation, eight consumer conditions of use are
evaluated based on the uses identified in EPA's 2015
TSCA Work Plan Chemical Problem Formulation and
Initial Assessment of 1.4-Dioxane (U.S. EPA. 2015). An
additional systematic review effort was undertaken for
consumer exposures to identify, screen, and evaluate
relevant data sources. These conditions of use include
use of 1,4-dioxane as a surface cleaner, antifreeze, dish
soap, dishwasher detergent, laundry detergent, paint and
floor lacquer, textile dye, and spray polyurethane foam
(SPF). 1,4-Dioxane may be found in these products at
low levels (0.0009 to 0.02%) based on its presence as a
byproduct of other formulation ingredients (i.e.,
ethoxylated chemicals). Inhalation exposures are
estimated for consumers and bystanders and dermal
exposures are estimated for users. Acute exposures are
presented for all consumer conditions of use, while
chronic exposures are presented for the conditions of use
that are reasonably expected to involve daily use
intervals (i.e., surface cleaner, dish soap, dishwasher
detergent, and laundry detergent). See Section 2.4.3 of
the final risk evaluation.

Page 67 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KPA/OPPT Response

SACC,
58

• The exclusion of subsurface and land disposal from the
Evaluation (Problem Formulation: page 44): leaves this
TSCA Evaluation incomplete. The Agency must assess the
concentrations of 1,4-Dioxane found in air and water
(surface and ground) near these injection facilities. This
determination cannot be made in the absence of such data.

• On page 209: The rationale for no further evaluation of the
disposal life stage seems to be tied to the comment in Table
B2 that states "2015 TRI data indicates 3 sites reporting
13,422 lbs to landfills. However, 1,4-Dioxane has low
sorption to soil." If 1,4-Dioxane is not sorbed to soils it
must be released as a vapor or transported to groundwater.
Both events produce risks that should be evaluated for
human and environmental health.

• Sludge or biosolids associated with disposed waste or
wastewater treatment facilities other than WWTPs must be
included in EPA's analysis of releases to land.

As described in Section 3.1 and Section 5.1.1, based on
its water solubility and log Kow 1,4-dioxane in soil,
sediment, and biosolids is expected to be in pore water
rather than sorbed to the solid fraction of the media.
Similarly, 1,4-dioxane released via surface or subsurface
disposal is not expected to sorb to soil but will migrate
to groundwater or surface water or, in relatively dry
conditions, will volatilize to air.

As described in Section 2.5.3.3 of the 1,4-dioxane
problem formulation (US EPA, 2018), ambient air,
drinking water, and ambient water exposure pathways
were not assessed in the 1,4-dioxane risk evaluation
because those media are addressed under the Clean Air
Act, Safe Drinking Water Act, and Clean Water Act.

SACC

The decision not to further analyze background levels of 1,4-
Dioxane in any matrix (Problem Formulation: page 47) cannot
be supported by any risk assessment principle. Any current use
scenarios increase exposures over those currently being
experienced.

Exposures pathways, including routes of exposure
through drinking water and ambient air, covered under
other EPA-administered statutes were not included in the
scope of the risk evaluation (see Section 1.4.2 of the
final risk evaluation).

r.nvironnienlal Kale Parameters

• The organic carbon-water partitioning coefficient (KOC)
has been estimated, not measured for 1,4-dioxane.
Justification is required for the decision to use the lower,
less conservative, KOC value when a more protective value

• There are two Koc-estimation methods included in
the EPI Suite™ KOCWIN module. The value
produced by the molecular connectivity index (MCI)
method was presented in the risk evaluation but is

Page 68 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

was also reported.

• An analysis of how the two values may affect uncertainty
in the predictive model outputs should be done.

o Using the lower value, EPA concluded that "1,4-
Dioxane is not expected to adsorb to soil and
sediment due to its low partitioning to organic
matter (estimated log KOC = 0.4)" (p. 212)

similar to the value estimated using the log Kow
method (log Koc = 0.4 by MCI and 0.6 by Kow).
The MCI method was presented in the risk
evaluation because it produced a slightly better fit to
measured values for the validation set (MCI r2 =
0.85, Kow method r2 = 0.78).

• The discrepancy in values is negligible and does not
produce uncertainty in the evaluation. Both values
lead to the conclusion that 1,4-dioxane will have low
potential to sorb to soil, sediment, or biosolids
particles.

Predictive modeling was used for environmental exposures
instead of available empirical environmental monitoring and
fate data. This is in violation of TSCA mandates.

• EPA identified only one study providing measured
values for environmental fate and transport.

• In instances where empirical data are not available,
uncertainty analysis must be performed to understand
the confidence that can be put into the models,
specifically soil partitioning modeling and fugacity
models, and to understand the impact of uncertainty and
variability on estimated risks.

• There are circumstances under which fugacity models
cannot accurately predict fate and transport of a
chemical such as 1,4-dioxane without empirical data or
more extensive modeling.

• For example, depending on groundwater flow and

• As described in the risk evaluation document,
systematic review of fate literature was undertaken
only for those environmental pathways and media
remaining in the environmental conceptual model
after the regulatory overlay was applied. Because
EPA determined during problem formulation that no
environmental pathways to ecological receptors
would be further analyzed in risk evaluation, EPA
limited data extraction and evaluation to key data
sources used in the 2015 EPA assessment of 1,4-
dioxane. The publication about biodegradation in
soil microcosms was the only publication used in
that risk assessment, so all other fate parameters
were estimated using EPI Suite™.

• The fugacity and STP removal models in EPI
Suite™ rely on p-chem properties for which

Page 69 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

hydrostatic conditions, there is some evidence that 1,4-
dioxane, when present in water as a contaminant, can
effectively be stored in place in the pore water and will
persist there. This is consistent with the conclusion in
the ATSDR Toxicological Profile that "1,4-Dioxane is
expected to persist in both water and soil."

measured values were collected (e.g., Henry's law
constant and vapor pressure) in conjunction with the
organic carbon-water partition coefficient (log Koc)
which was estimated using a molecular connectivity
index (MCI) method with high reliability. EPI
Suite™ also includes a method of estimation log Koc
based on log Kow, and the result of that estimation
method is similar to the MCI method result. Thus,
although measured values were not obtained for
some fate endpoints, the uncertainty surrounding
fugacity model results is relatively low.

• The persistence of 1,4-dioxane in water and soil as
described by ATSDR is not inconsistent with the
results of the fugacity models in EPI Suite™. The
summary statement about the fate of 1,4-dioxane
(Section 3.1, pg. 52) was modified to clarify the
expected environmental persistence of 1,4-dioxane.

r.nvironnienlal Kisk Characterization

24, 58

The words "conservative approach" and "conservative
assumptions" are inappropriately associated with assessment
factors in developing aquatic COCs. Assessment factors cannot
be construed as "safety factors" that yield conservative
estimates.

EPA has revised the hazard assessment section of the
risk evaluation. The updated section includes a weight-
of-evidence approach for selecting the most relevant
species for the surface water environment. The receptor
that is the most relevant on the population level for
short-term exposure to 1,4-dioxane in the aquatic
environment is the algal endpoint. EPA has modified the
calculations for deriving the COC for this endpoint. The
COC for algae is derived by dividing the hazard value by
an assessment factor of ten. For the chronic endpoint, the
most relevant species on the population level is the fish

Page 70 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

endpoint. The method for deriving the chronic COC has
not changed.

EPA acknowledges that there are some limitations in
applying assessment factors for certain species.
However, EPA stands by the analysis that was
conducted to determine the environmental hazard,
exposure and risk of 1,4-dioxane to the environment.
After a complete analysis of the hazard data and
exposure assessment of 1,4-dioxane, EPA is confident
that the risks of this chemical are low to aquatic
organisms. These conclusions are supported by other
countries that have investigated the risk of 1,4-dioxane.

• EPA compares both monitoring data and modeling data
predicting surface water concentrations near discharging
facilities to the acute and chronic COCs to generate risk
quotients. Why does EPA need to use both modeling and
monitoring data?

o EPA may want to consider developing guidance or
a flow chart describing how both monitoring and
modeling data should be handled to inform a tiered
approach to assessment.

• The above comparison should be re-done using the
corrected acute COC calculated from the lowest toxicity
value in Table F-l (575 mg/L).

• Section 4.1.1 of the draft risk evaluation only utilizes
the predicted surface water levels obtained from the
first-tier aquatic exposure screen. Because it was
decided during problem formulation based on this
first-tier assessment not to further analyze this
pathway, EPA utilized modeled data derived from
known releasers to derive risk quotients as part of the
risk characterization.

• EPA inadvertently calculated the short term (acute)
concentrations of concerns (COCs) for fish rather
than for the algal endpoint. EPA has since updated
the COC for the correct endpoint. In addition, EPA
has updated the calculations used for determining the
assessment factor for the acute endpoint.

24, 58

With little analysis and based on limited data, EPA asserts that
"[mjeasured and estimated levels of 1,4-dioxane in the

• EPA has revised the hazard assessment section of the
risk evaluation. The updated section includes a

Page 71 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KIW/OPPT Response

environment are sufficiently below the acute and chronic
aquatic COCs [concentrations of concern]," plans no further
analysis, and implies it has concluded that any associated risks
can be ignored (p. 41). Yet:

• EPA's predicted concentrations in surface water for acute
and chronic scenarios are up to 58% and 40% of the COCs,
leaving little room for error.

• EPA implies that its calculations of COCs are conservative
at least in part because of its use of assessments factors (pp.
29, 70, 81). The use of such factors is not conservative:
They account for real-world sources of variability as well
as database limitations, and cannot be construed as "safety
factors" that yield conservative estimates. 158 As EPA
states: "The application of AFs [assessment factors]
provides a lower bound effect level that would likely
encompass more sensitive species not specifically
represented by the available experimental data. AFs are
also account for differences in inter- and intra-species
variability, as well as lab oratory-to-fi eld variability." (p.

70)

• EPA's calculated acute COC is inconsistently reported. In
the text, it is listed as 59,800 ppb (p. 35), while in
Appendix C it is listed as 20,000 ppb (p. 70).

weight-of-evidence approach for selecting the most
relevant species for the surface water environment.
The species that is the most relevant on the
population level for short-term exposure to 1,4-
dioxane in the aquatic environment is the algal
endpoint. EPA has modified the calculations for
deriving the COC for this endpoint. The COC is for
algae is derived by dividing the hazard value by an
assessment factor of ten instead of four. For the
chronic endpoint, the most relevant species on the
population level is the fish endpoint. The method for
deriving the chronic COC has not changed. The
assessment factor is ten.

• EPA stands by the analysis that was conducted to
determine the environmental hazard and risk of 1,4-
dioxane to the environment. After a complete
analysis of the hazard data of 1,4-dioxane and the
quantitative and qualitative exposure analyses
conducted during problem formulation, EPA is
confident that the risks of this chemical are low to
aquatic organisms. These conclusions are supported
by other countries that have investigated the hazard
of 1,4-dioxane.

Page 72 of 212

Summary of Comments for Specific
Issues Related to Charge Question 3

KPA/OPPT Response

assessment factor for the acute endpoint. The
assessment factor is ten (10).

4. Exposure and Releases

Charge Question 4.1: Please comment on the characterization of occupational inhalation exposure for workers for each of the
identified conditions of use. What other additional information, if any, should be considered?

Charge Question 4.2: Please comment on the characterization of occupational inhalation exposure for occupational non-users for
each of the identified conditions of use. What other additional information, if any, should be considered?

Charge Question 4.3: Please comment on the characterization of occupational dermal exposure for workers. What other additional
information, if any, should be considered?

Charge Question 4.4: Please comment on the approach for characterizing the different use scenarios. Are there any additional 1,4-
dioxane specific data and/or information that should be considered?

Summary of Comments for Specific
Issues Related to Charge Question 4

KPA/OPPT Response

SACC

• Include members on the assessment team with a strong
industrial hygiene background relative to modeling and
the setting and comparing of occupational exposure
limits to estimated levels of potential exposure.

• The EPA multidisciplinary team includes an industrial
hygienist who has reviewed and/or contributed to this risk
evaluation document.

SACC

• Add more information concerning the context (e.g.,
measurement and methodology details) of monitoring
data used in the Evaluation.

• EPA provides reasonable context for the data in the
current risk evaluation. Evaluation criteria can be found
in the supplemental document titled Application of
Systematic Review in TSCA Risk Evaluations. The scores
and rationale for each score can be found in the
supplemental document titled Data Quality Evaluation of
Environmental Releases and Occupational Exposure

Page 73 of 212

-------




Data. Monitoring test protocols and results were assessed
for quality of data, including quality of measurement and
methodology.

SACC

• Obtain, using TSCA authority, additional monitoring
and specific scenario data on workers and ONUs
linked to the specific drivers within the scenarios
causing that exposure that would help to reduce
uncertainties associated with this assessment.

EPA used reasonably available information to construct
exposure scenarios for workers and ONUs for the conditions
of use of 1,4-dioxane. Assumptions regarding worker and
ONU exposures were discussed in Section 2.4.1. Additional
clarifications are also included:

Workers that are directly handling 1,4-dioxane and/or
perform activities near sources of 1,4-dioxane are in the near
field and are called workers throughout this risk evaluation.
The near-field is defined as a volume of air within one-meter
in any direction of the worker's head and the far-field
comprised the remainder of the room {Tielemans, 2008,
2599270}. The source areas/exposure zones are
conceptualized and delineated by several factors such as the
quantity of 1,4-dioxane releases, ventilation of the facility,
vapor pressure and emission potential of the chemical,
process temperature, size of the room, job tasks, and modes
of chemical dispersal from activities {Leblanc, 2018,
4140533}. Corn and Esmen {1979, 29525} indicated that the
assignment of zones is a professional judgment and not a
scientific exercise.

SACC

• The hierarchy for breathing zone exposure potential
should be reformulated to put modeling ahead of
monitoring for poorly described scenarios.

EPA used model and relevant parameter data for
Occupational Exposure Scenarios (OESs) for inhalation
(personal breathing zones) and dermal exposure conditions
as listed in Table 4-13. EPA did not find reasonably available
data for modeling of breathing zones for other OESs. In
addition, the breathing zone exposure models cannot be
validated using the full range of possible exposure
concentrations.

SACC

• The steady-state breathing-zone concentration model
used by the Agency for interior rooms should be

• Computational fluid dynamics or other advanced models
could perform better simulation of airflow and transport of

Page 74 of 212

-------


discarded as out-of-date. A team member (see #2
above) with knowledge of contemporary AIHA models
should handle the modeling in this document.

1,4-dioxane in occupied spaces. To capture these
behaviors with a high degree of accuracy requires spatial
(grid) resolution and time step refinement that can directly
simulate appropriate scenario. This was certainly not
achieved here, as the simulation could not be validated via
tests data. Model results work well or poorly depending
on data and known constraints for their applications, even
when the application is limited to a point source in a room
under general ventilation. Numerous researchers,
including National Research Council (see Using Modeling
and Simulation in Test Design and Evaluation by Panel on
Statistical Methods for Testing and Evaluating Defense
Systems, 1998. DOI: https://doi.org/10.17226/6037),
agreed that no simulations, experiments, or field tests will
fully represent the universe of possible scenarios to which
these models could be reasonably applied by industrial
hygienists. This is a limitation shared by scientists for any
method of model evaluation. Inhalation exposures in the
current assessment were evaluated using modeled data
along with personal breathing zone (PBZ) samples.
Assumptions of steady-state air concentrations were
necessary and associated with the development of this risk
evaluation. EPA may consider the use of alternate and
more sophisticated modelling in conjunction with
validated monitoring data in future risk evaluation work.

SACC

• Add a spill scenario to this assessment.

Spills and leaks generally are not included within the scope of
a TSCA risk evaluation because in general they are not
considered to be circumstances under which a chemical
substance is intended, known or reasonably foreseen to be
manufactured, processed, distributed, used, or disposed of. To

Page 75 of 212

-------
the extent there may be potential exposure from spills and
leaks, EPA is also declining to evaluate environmental
exposure pathways addressed by other EPA-administered
statutes and associated regulatory programs.

First, EPA does not identify 1,4-dioxane spills or leaks as
"conditions of use." EPA does not consider 1,4-dioxane spills
or leaks to constitute circumstances under which 1,4-dioxane
is manufactured, processed, distributed, used, or disposed of,
within TSCA's definition of "conditions of use." Congress
specifically listed discrete, routine chemical lifecycle stages
within the statutory definition of "conditions of use" and EPA
does not believe it is reasonable to interpret "circumstances"
under which 1,4-dioxane is manufactured, processed,
distributed, used, or disposed of to include uncommon and
unconfined spills or leaks for purposes of the statutory
definition. Further, EPA does not generally consider spills and
leaks to constitute "disposal" of a chemical for purposes of
identifying a COU in the conduct of a risk evaluation.

In addition, even if spills or leaks of 1,4-dioxane could be
considered part of the listed lifecycle stages of 1,4-dioxane,
EPA has "determined" that spills and leaks are not
circumstances under which 1,4-dioxane is intended, known or
reasonably foreseen to be manufactured, processed,
distributed, used, or disposed of, as provided by TSCA's
definition of "conditions of use," and EPA is therefore
exercising its discretionary authority under TSCA Section
3(4) to exclude 1,4-dioxane spills and leaks from the scope of
the 1,4-dioxane risk evaluation. The exercise of that authority
is informed by EPA's experience in developing scoping
documents and risk evaluations, and on various TSCA
provisions indicating the intent for EPA to have some	

Page 76 of 212

-------
discretion on how best to address the demands associated with
implementation of the full TSCA risk evaluation process.
Specifically, since the publication of the Risk Evaluation
Rule, EPA has gained experience by conducting ten risk
evaluations and designating forty chemical substances as low-
and high-priority substances. These processes have required
EPA to determine whether the case-specific facts and the
reasonably available information justify identifying a
particular activity as a "condition of use." With the experience
EPA has gained, it is better situated to discern circumstances
that are appropriately considered to be outside the bounds of
"circumstances... under which a chemical substance is
intended, known, or reasonably foreseen to be manufactured,
processed, distributed in commerce, used, or disposed of' and
to thereby meaningfully limit circumstances subject to
evaluation. Because of the expansive and potentially
boundless impacts that could result from including spills and
leaks as part of the risk evaluation, (e.g., due to the
unpredictable and irregular scenarios that would need to be
accounted for, including variability in volume, frequency, and
geographic location of spills and leaks; potential application
across multiple exposure routes and pathways affecting
myriad ecological and human receptors; and far-reaching
analyses that would be needed to support assessments that
account for uncertainties but are based on best available
science),_which could make the conduct of the risk evaluation
untenable within the applicable deadlines, spills and leaks are
determined not to be circumstances under which 1,4-dioxane
is intended, known or reasonably foreseen to be
manufactured, processed, distributed, used, or disposed of, as
provided by TSCA's definition of "conditions of use."

Exercising the discretion to not identify spills and leaks of

Page 77 of 212

-------
1,4-dioxane as a COU is consistent with the discretion
Congress provided in a variety of provisions to manage the
challenges presented in implementing TSCA risk evaluation.
See e.g., TSCA Sections 3(4), 3(12), 6(b)(4)(D), 6(b)(4)(F).
In particular, TSCA Section 6(b)(4)(F)(iv) instructs EPA to
factor into TSCA risk evaluations "the likely duration,
intensity, frequency, and number of exposures under the
conditions of use...suggesting that activities for which
duration, intensity, frequency, and number of exposures
cannot be accurately predicted or calculated based on
reasonably available information, including spills and leaks,
were not intended to be the focus of TSCA risk evaluations.
And, as noted in the preamble to the Risk Evaluation Rule,
EPA believes that Congress intended there to be some
reasonable limitation on TSCA risk evaluations, expressly
indicated by the direction in TSCA Section 2(c) to "carry out
[TSCA] in a reasonable and prudent manner."

For these reasons, EPA is exercising this discretion to not
consider spills and leaks of 1,4-dioxane to be COUs.

Second, even if 1,4-dioxane spills or leaks could be identified
as exposures from a COU in some cases, these are generally
not forms of exposure that EPA expects to consider in risk
evaluation. TSCA Section 6(b)(4)(D) requires EPA, in
developing the scope of a risk evaluation, to identify the
hazards, exposures, conditions of use, and potentially exposed
or susceptible subpopulations the Agency "expects to
consider" in a risk evaluation. This language suggests that
EPA is not required to consider all conditions of use, hazards,
or exposure pathways in risk evaluations. EPA has chosen to
tailor the scope of the risk evaluation to exclude spills and
leaks in order to focus analytical efforts on those exposures

Page 78 of 212

-------
that present the greatest potential for risk.

In the problem formulation documents for many of the first 10
chemicals undergoing risk evaluation, EPA applied the same
authority and rationale to certain exposure pathways,
explaining that "EPA is planning to exercise its discretion
under TSCA 6(b)(4)(D) to focus its analytical efforts on
exposures that are likely to present the greatest concern and
consequently merit a risk evaluation under TSCA...." This
approach is informed by the legislative history of the
amended TSCA, which supports the Agency's exercise of
discretion to focus the risk evaluation on areas that raise the
greatest potential for risk. See June 7, 2016 Cong. Rec.,
S3519-S3520.

In addition to TSCA Section 6(b)(4)(D), the Agency also has
discretionary authority under the first sentence of TSCA
Section 9(b)(1) to "coordinate actions taken under [TSCA]
with actions taken under other Federal laws administered in
whole or in part by the Administrator." TSCA Section 9(b)(1)
provides EPA authority to coordinate actions with other EPA
offices, including coordination on tailoring the scope of
TSCA risk evaluations to focus on areas of greatest concern
rather than exposure pathways addressed by other EPA-
administered statutes and regulatory programs, which does
not involve a risk determination or public interest finding
under TSCA Section 9(b)(2). EPA has already tailored the
scope of this risk evaluation using such discretionary
authorities with respect to exposure pathways covered under
the jurisdiction of other EPA-administered statutes and
associated regulatory programs (see section 1.4.3).

Following coordination with EPA's Office of Land and	

Page 79 of 212

-------




Emergency Management (OLEM), EPA has found that
exposures of 1,4-dioxane from spills and leaks fall under the
jurisdiction of RCRA. See 40 CFR 261.33(d) (defining in part
a hazardous waste as "any residue or contaminated soil, water
or other debris resulting from the cleanup of a spill into or on
any land or water of any commercial chemical product or
manufacturing chemical intermediate having the generic name
listed [40 CFR 261.33(e) or (f)], or any residue or
contaminated soil, water or other debris resulting from the
cleanup of a spill, into or on any land or water, of any off-
specification chemical product and manufacturing chemical
intermediate which, if it met specifications, would have the
generic name listed in [40 CFR 261.33(e) or (f)]"); 40 CFR
261.33(f) (listing 1,4-dioxane as hazardous waste no. U108).
As a result, EPA believes it is both reasonable and prudent to
tailor the TSCA risk evaluation for 1,4-dioxane by declining
to evaluate potential exposures from spills and leaks, rather
than attempt to evaluate and regulate potential exposures from
spills and leaks under TSCA. See Section 2.4.1 in the final
risk evaluation for further explanation on a COU basis.

SACC

• Add a fugitive emissions scenario to this assessment.

EPA did not include the emission pathways to ambient air
from commercial and industrial stationary sources, because
stationary source releases of 1,4-dioxane to ambient air are
under the jurisdiction of and addressed by Section 112 of the
Clean Air Act (CAA). The resulting exposure pathways were
out of scope as described in Section 1.4.2.

SACC

• Add scenarios in which respirators are not used for an
entire 8-hour work shift.

The risk evaluation already presents exposure and risk
estimates with and without PPE, which includes no
respirators. As previously noted, 1,4-dioxane is manufactured,
processed, and used in industrial settings, where there are
typically strong industrial hygiene programs that include
training and oversight. It is not reasonable to assume no
respirator use in these settings. For situations in which

Page 80 of 212

-------




workers do not use respirators for an entire shift, EPA used
the high-end exposure value when considering worker risks in
order to address the uncertainties and variability in PPE usage

SACC

• Consider scenarios in which acute exposures occur on
time frames of less than 8 hours.

Exposures at different facilities could occur for a variety of
conditions of use and those scenarios are discussed in the risk
evaluation document based on the reasonably available
information. EPA assessed occupational exposure scenarios
less than 8-hour as well as full shift depending on conditions
of use. Examples of less than 8-hour work shift includes 15-
minute TWA (Evaporator Dump) for manufacturing, short
(15-minute) duration degassing for laboratory use; and the
30-minute exposure for repackaging into drums.

SACC

• Present worst-case inhalation exposures for workers
with estimates from scenarios assuming no use of PPE.

The risk evaluation presents exposure and risk estimates with
no use of PPE as worst-case for various occupational
exposure scenarios.

SACC

• In addition to a qualifier for the quality of data used for
estimating inhalation exposures, EPA should add a
qualifier for the overall confidence in the final
exposure estimates (in addition to the description of
uncertainties).

The quality and quantity of monitoring data, surrogate,
modeling and other information were listed in Table 2-19.
The Table 2-19 also provides confidence ranking for various
COUs.

SACC

• Clarify how ONU exposures compare to both
unprotected and protected user exposures.

EPA considers occupational non-users (ONUs) to be a subset
of workers for whom the potential inhalation exposures may
differ based on proximity to the exposure source. For the
majority of 1,4-dioxane conditions of use, the difference
between ONU exposures and workers directly handling the
chemical cannot be quantified. EPA assumed an absence of
PPE for ONUs, since ONUs do not directly handle the
chemical and are instead doing other tasks in the vicinity of
1,4-dioxane use. EPA also assumed that, in most cases, ONU
inhalation exposures are lower than inhalation exposures for
workers directly handling the chemical substance. For dermal
exposures, because ONUs are not dermally exposed to 1,4-

Page 81 of 212

-------




dioxane, dermal risks to ONUs were not identified. For
inhalation exposures, EPA, where possible, estimated ONU
exposures and described the risks separately from workers
directly exposed. To account for those instances where, based
on EPA's analysis, the monitoring data or modeling data for
worker and ONU inhalation exposure could not be
distinguished, EPA considered the central tendency risk
estimate when determining ONU risk.

SACC

• Engage an expert in dermal exposure assessment from
within the Agency or a consultant to provide
quantitative estimates of the amount of 1,4-Dioxane
absorbed systemically in reasonably anticipated
scenarios.

• Dermal exposure assessments have been updated by using
sensitivity analysis; and comparing test results performed
at different facilities. Peer-reviewed references are cited as
appropriate.

SACC

• Strengthen the discussion and analysis for uncertainty
for dermal exposure by quantitatively defining the
assumptions made in each scenario using a Monte
Carlo simulation.

• Monte Carlo simulation and other sensitivity analysis
have been performed as applicable for the inhalation and
dermal exposures. The operating variables and constraints
for those conditions along with the assumptions, citations
of the references for the data used are already included in
the risk evaluation.

SACC

• Clearly state the estimated exposures to gloved and
ungloved Users.

The risk evaluation presents exposure and risk estimates for
workers with and without PPE, including gloves. As
previously noted, ONUs do not handle the chemical;
therefore, dermal exposures were not assessed for ONUs.

SACC

• Contrast ONU estimated exposures to estimated
exposures of gloved and ungloved Users.

EPA has updated several sub-sections in Section 4.1.4 to
show risk estimates for workers with PPE and without PPE,
and risk estimates for ONUs, who EPA assumes do not have
PPE. As previously noted, ONUs do not handle the chemical;
therefore, dermal exposures were not assessed for ONUs.

SACC

• Resolve the large discrepancy between the theoretical
predictions of high dermal doses and apparently low
systemic uptake as reported by experimental

As indicated earlier in this response to comment document,
the dermal exposure calculations have been updated and
validated by using sensitivity analysis and by comparing test
results reported by the researchers at the Kansas State
University, Manhattan, Kansas; and at the University

Page 82 of 212

-------


observations.

Erlangen-Nuremberg, Erlangen, Germany, and the citations
are included in the revised risk evaluation document (see
Section 2.4.1.14 - Dermal Exposure Assessment).

SACC

• Conduct in vitro testing (OECD 2004) of the dermal
absorption of 1,4-Dioxane.

The peer-reviewed publications that reported test results (in-
vivo and ex-vivo) using 1,4-dioxane and human skin are
discussed, as appropriate, in the revised dermal exposure
assessments of the risk evaluation. The researchers at the
Kansas State University, Manhattan, Kansas and at the
University Erlangen-Nuremberg, Erlangen, Germany
performed in-vitro and ex-vivo dermal absorptions using 1,4-
dioxane and these results with interpretations are included in
the Section 2.4.1.14 (Dermal Exposure Assessment) with
citations.

SACC

• Define scenarios as exposure settings that have a
comprehensive set of the determining factors that
cause the exposure. When these are matched to
monitoring data, it is a very powerful tool for assessing
exposure and risk. Without monitoring data, models
should be used with these scenarios.

Modeling and monitoring data were used to compare the
validity of the assessment, as appropriate. When model
calibration data and monitoring data needed for performance
evaluation and validation of mathematical models are not
available, the limitations are recognized, assumptions are
specified, and literature data (or surrogate data) are identified
during the interpretation of scenarios.

24,
30,
43,
50, 55
58

Consumer Exposure

•	EPA claims it "did not find evidence of any current
consumer uses for 1,4-dioxane and is excluding
consumer uses from the scope of the risk evaluation."
"Contamination of industrial, commercial and
consumer products or presence as a byproduct are not
intended conditions of use for 1,4-dioxane and
therefore will not be evaluated."

•	EPA must account for this pathway of exposure now
as part of the cumulative exposure to the general
population and workers.

•	Exclusion of evaluation of exposure from consumer
uses is in violation of TSCA requirements.

As explained in the scope document, 1,4-dioxane may be
found as a contaminant in consumer products that are readily
available for public purchase. In the final risk evaluation,
eight consumer conditions of use are evaluated based on the
uses identified in EPA's 2015 TSCA Work Plan Chemical
Problem Formulation and Initial Assessment of 1,4-Dioxane
( ). An additional systematic review effort was
undertaken for consumer exposures to identify, screen, and
evaluate relevant data sources. These conditions of use
include use of 1,4-dioxane as a surface cleaner, antifreeze,
dish soap, dishwasher detergent, laundry detergent, paint and
floor lacquer, textile dye, and spray polyurethane foam (SPF).
Regarding body care and cosmetic products, they are

Page 83 of 212

-------
Many consumer uses were correctly recognized in the
scoping document, but not evaluated.

1,4-dioxane as a residual contaminant has been
detected in paints, coatings, lacquers, ethylene glycol-
based antifreeze coolants, spray polyurethane foam,
household detergents (dish and laundry),
cosmetics/toiletries including shampoo, body wash,
baby wipes, soaps, lotions, sunscreen and toothpaste,
textile dyes, clothing, baby bibs, blankets/throws, bath
and pool water toys, pharmaceuticals, foods,
agricultural and veterinary products, magnetic tape and
adhesives, and in compost sold for home and garden
use.

The justification for excluding consumer uses based on
presence of 1,4-dioxane as a "byproduct" or a
contaminant, rather than an intentional use, is
unacceptable.

TSCA does not distinguish between intentional and
unintentional presence of a chemical in a product.
Deliberate uses reported in the 2002 EU risk
assessment were ignored.

For cleaning agents and paint as end uses, the EU risk
assessment found reasonable worst case to be 50
mg/m3 and the typical concentration to be 15 mg/m3,
which are considerably higher than the Central
Tendency average daily concentration (ADC) and
High-End ADCs EPA relies on for all of its exposure
scenarios (see Table 5-5 on p. 137).

According to the reports submitted to FracFocus, 1,4-
dioxane is not just present as an impurity or by-
product. Companies reported over 400 instances where
1,4-dioxane was used as an ingredient, representing
77% of the reported cases in this time period.	

excluded from the definition of "chemical substance" per
TSCA section 3(2) and are outside the scope of this risk
evaluation. 1,4-Dioxane may be found in these products at
low levels (0.0009 to 0.02%) based on its presence as a
byproduct of other formulation ingredients {i.e., ethoxylated
chemicals). Inhalation exposures are estimated for consumers
and bystanders and dermal exposures are estimated for users.
Acute exposures are presented for all consumer conditions of
use, while chronic exposures are presented for the conditions
of use that are reasonably expected to involve daily use
intervals {i.e., surface cleaner, dish soap, dishwasher
detergent, and laundry detergent). See Section 2.4.3 of the
final risk evaluation.

• EPA has reviewed the list of FracFocus reports on 1,4-
dioxane submitted by the commenter as well as the three
individual job reports submitted by the commenter. In
one individual job report, the 1,4-dioxane is specifically
noted as an impurity. In the other two, the percentage of
1,4-dioxane reported was low, indicating that the 1,4-
dioxane is likely present as an impurity in the ethoxylated
alcohols that are also named in the same reports. EPA
initially excluded production of 1,4-dioxane as a by-
product from certain other chemicals and presence as a
contaminant in industrial, commercial and consumer
products from the scope of the risk evaluation using
EPA's discretion under TSCA section 6(b)(4)(D). While
EPA has addressed some conditions of use related to 1,4-
dioxane as a byproduct in this risk evaluation, EPA
expects that 1,4-dioxane exposures associated with the
use of ethoxylated alcohols used in hydraulic fracturing
fluids would be considered in the scope of a risk
evaluation for ethoxylated alcohols. In cases like this,

Page 84 of 212

-------


•	The choice to exclude 1,4-dioxane's presence as a
byproduct as a condition of use in the risk evaluation
provides no basis for EPA not to include 1,4-dioxane's
presence as an intentionally added substance.

•	1,4-Dioxane has been detected in landfill leachate
(ATSDR, 2012); this suggests that it may be coming
from consumer products containing it.

EPA believes its regulatory tools under TSCA section
6(a) are better suited to addressing any unreasonable risks
that might arise from these activities through regulation
of the activities that generate 1,4-dioxane as an impurity
or cause it to be present as a contaminant than they are to
addressing them through direct regulation of 1,4-dioxane.
This case-by-case approach for byproducts exposures is
consistent with the various scenarios explained in the
Risk Evaluation Rule, 82 FR at 33730.

As described in Section 1.4.2 of the risk evaluation, EPA
believes it is both reasonable and prudent to tailor TSCA risk
evaluations when other EPA offices have expertise and
experience to address specific environmental media, including
landfill leachate under RCRA, rather than attempt to evaluate
and regulate potential exposures and risks from those media
under TSCA. EPA has therefore tailored the scope of the risk
evaluation for 1,4-dioxane using authorities in TSCA Sections
6(b) and 9(b)(1).

24, 58

Occupational/Workplace Exposure

Reliance on Limited or Questionable Data

•	Overall, EPA's sources of workplace exposure data are
from selective, unrepresentative sources; lack critical
detail on which processes, exposure sources and
worker activities they represent; are insufficient to
understand the distribution of exposures in a given
setting; and reliable occupational exposure data are
limited. EPA should require the production of
reasonable exposure data.

•	Oral exposure should be assessed as an exposure
pathway.

•	For its Manufacturing scenario, EPA chose to use data
it received from BASF, comprised of 30 samples from

•	EPA received monitoring data from industry consortium
(Halogenated Solvents Industry Alliance, Inc., Arlington,
Virginia), Department of Defense, National Institute for
Occupational Safety and Health (NIOSH), Occupational
Safety and Health Administration (OSHA), and Kansas
City National Security Campus (KCNSC), Kansas City,
Missouri. These monitoring data were discussed and
interpreted in the risk evaluation and supporting
documents.

•	There are two known manufacturers in the U.S., and one
of these sites is representative of the U.S. manufacturing.
One of the two manufacturers (BASF) provided
information that is more relevant and recent compared in
comparison to other source that lacks monitoring data (see

Page 85 of 212

-------
a single manufacturing site, and assumed these data to
be representative of all U.S. manufacturing. The data
also lacked specific descriptions of worker tasks,
exposure sources, and possible engineering controls to
provide context.

o EPA assumed that the 2016 BASF data are
PBZ measurements relevant to worker
activities and are also 8-hour TWA
measurements. This assumption could
underestimate exposures. The sampling rate
was missing for some of the 2016 data, so EPA
assumed the same sampling rate was applied
for other data in the set. These assumptions
were not included in the "assumptions and key
sources of uncertainty section."
o It was also noted that access to the original data
source was not provided, and no link was
provided for the HERO entry.

•	Regarding open system functional fluids: EPA claims
it derived fluid concentrations from available SDSs (p.
62), but none of the relevant cited SDSs that are
publicly accessible makes any mention of 1,4-dioxane
as a constituent. EPA's cited source (Burton and
Driscoll 1997) is aNIOSH site report, motivated by
worker concern over fungi- and bacteria- contaminated
synthetic metal-working fluids (MWF). It entailed no
direct measurements of 1,4-dioxane, only synthetic
MWF and it is not clear the fluids at this site even
contained the chemical (p. 61).

•	Regarding printing inks: EPA's analysis of worker
exposure to printing inks is based on a single air
sample reported in a 2016papery despite the fact that
the authors and other researchers note that the

details in Section 4.1.4 of the risk evaluation). EPA
updated additional discussion about considerations made
in the Assumptions and Key Sources of Uncertainty
section in response to sub-comment 3a. EPA has
additionally corrected the HERO entry for the BASF,
2016 citation (HERO ID 5079874) so that the original
data source can be accessed.

•	Citations with appropriate references including Burton
and Driscoll (1997) are detailed in Appendix G of the risk
evaluation. Though Burton and Driscoll (1997) do not
address 1,4-dioxane in metalworking fluids (MWF), they
indicated "PBZ and area measurements of water-soluble
synthetic metalworking fluids and oil mists from
conventional metalworking fluids.'" This pertinent
information along with the SDS information were used to
address potential exposure to 1,4-dioxane from metal
working fluids.

•	The "basis for single air sample reported in a 2016 paper"
is provided in the Key Uncertainties section:

"Additionally, the sample provided is not a PBZ sample.
Since the sample was taken within the 3D printing
enclosure, the exposure value is likely higher than a
worker would typically experience while operating the 3D
printer."

•	Regarding the Industrial Uses related "explanation for
considering highest exposure point", EPA provides the
explanation for considering the 184mg/m3 datapoint as an
outlier in the Appendix G. In response to this comment,
EPA further clarified the explanation in the Appendix G -
Section 6.3).

•	Regarding "1,4-dioxane mists," Table G-17, 1997NIOSH
HHE PBZ and Area Sampling Data for Metalworking
Fluids in Appendix G of the risk evaluation provides the

Page 86 of 212

-------


concentration could well be an underestimate (p. 70),
EPA asserts it is likely an overestimate (p. 71). The
basis for this assertion should be provided.

•	For Industrial Uses, EPA excludes the highest
exposure point (184 mg/m3) from the 2002 EU Risk
Assessment. An explanation for considering this value
"likely an outlier" (p. 264) should be provided.

•	EPA draws the conclusion that exposure to 1,4-
dioxane mists would be negligible; supporting analysis
or data should be provided.

•	An available workplace monitoring study (OSHA
2016) was inappropriately scored as unacceptable,
EPA should attempt to collect this potentially useful
data.

complete rationale and approach for determining worker
exposure to 1,4-dioxane from mists generated during
metalworking. EPA used data from the analysis outlined
here to develop the conclusions on 1,4-dioxane mists.

• EPA converted the OSHA 2016 document into a file
format that was legible and re-examined the data. The
data is OSHA sampling data from 2016 and all of the
entries are non-detects. As described by Hornung and
Reed (1990), the method for estimating non-detects
depends upon the degree to which the data are skewed and
the proportion of the data that is below detection limits. If
the geometric standard deviation of the monitoring data
set is less than 3.0, nondetectable values should be
replaced by the limit of detection divided by the square
root of two (L/V2).If the data are highly skewed, with a
geometric standard deviation of 3.0 or greater,
nondetectable values should be replaced by half the
detection limit (L/2). The scoring for the document has
been changed from unacceptable to medium. Per OSHA
recommendation, EPA did not use non-detects from
OSHA (2016). The uncertainties of OSHA (2016) data
quality has been also recognized due to the lack of clarity
- whether the source areas from where the samples were
collected contained 1,4-dioxane.

21,24
46,
48,
55,
58,

Use of PPE

EPA assumes that PPE equipment is a) provided to all
workers, b) used by all workers (properly), and c) that PPE
equipment used would provide adequate protection such
that there are few unreasonable risks to workers. Support
for these assumptions should be provided.

• OSHA previously informed EPA that it "considers]
the use of respirators to be the least satisfactory
approach to exposure control." EPA has not published

• The quote from the commenter does not reflect OSHA
review of this risk evaluation; rather, it cites to a comment
on a 2014 proposed SNUR. OSHA participated in review
of EPA's draft risk evaluation and final risk evaluation of
1,4-dioxane. EPA has recognized in the draft and final
risk evaluation documents regarding OSHA's hierarchy of
controls and recognized that there can be reliability issues
associated with PPE. EPA's risk evaluation characterizes
risks with and without PPE considerations, with

Page 87 of 212

-------
the results of OSHA's review of the draft 1,4-dioxane
risk evaluation.

EPA assumes that the three general OSHA standards -
the Personal Protective Equipment Standard, the
Respiratory Protection Standard, and the Hazard
Communication Standard - will ensure that workers
have access to and use appropriate PPE.
EPA overstates OSHA's authorities and requirements,
claiming that OSHA requires employers to provide
PPE (p. 48), requires the use of respirators for 1,4-
dioxane (p. 52), and that the OSHA requirement for
safety data sheets (SDSs) is sufficient to ensure use of
protective measures such as PPE by all downstream
users of 1,4-dioxane (p. 60).

OSHA regulations do not require that persons comply
with SDSs and rely heavily on employers to adhere to
other requirements.

OSHA does not require employers to use the same
regulatory thresholds as EPA.

EPA made no attempts to coordinate with any of these
other statutes to assure measures to mitigate risk were
being taken.

If a chemical presents a significant risk, OSHA and
NIOSH manage that risk using the "hierarchy of
controls," under which hazard elimination,
substitution, engineering and administrative controls
are all prioritized over the use of PPE. Assuming PPE
use ignores this hierarchy.

Under section 5 of the OSH Act, employers are
required to comply with OSHA standards and provide
a workplace free from recognized hazards. However,
employers have no obligation to protect workers from
chemical hazards that are not the subject of an OSHA

considerations of engineering and administrative controls.
Additionally, in consideration of the uncertainties and
variabilities in PPE usage, EPA uses the high-end
exposure value when making its unreasonable risk
determination in order to address those uncertainties. EPA
has also outlined its PPE assumptions in section 5.1.
Information reasonably available to EPA, including data
submitted by chemical manufacturers and processors,
indicates that PPE is generally used. EPA does not assume
that inclusion of PPE on SDSs is sufficient to ensure PPE
use and while EPA considers the information on SDSs,
EPA does not make PPE use assumptions based solely on
SDSs. EPA is not yet at the risk management stage for
this chemical, where the hierarchy of controls would be
considered. OSHA's hierarchy of controls is a method for
eliminating workplace hazards. While EPA has assessed
the extent to which certain exposure reduction tools that it
assumes to be in place may be reducing risks to workers,
application of the methodology of the hierarchy of
controls is not relevant to risk evaluations. EPA will
manage unreasonable risks presented by chemical
substances when the Agency undertakes regulatory action
for COUs determined to have unreasonable risk.
Utilization of the hierarchy of controls to recommend or
require risk management actions in the risk evaluation
would be premature and inappropriate.

1,4-Dioxane is the subject of an OSHA standard. OSHA
has established a permissible exposure limit (PEL) of 100
ppm (8-hour TWA) for 1,4-dioxane. However, as noted
on OSHA's website, "OSHA recognizes that many of its
permissible exposure limits (PELs) are outdated and
inadequate for ensuring protection of worker health. Most

Page 88 of 212

-------
standard unless the employer has actual knowledge
that the chemical poses a risk to workers. OSHA has
never recognized a government risk assessment as the
basis for finding that an employer has actual
knowledge of a chemical hazard, and the standards for
evaluating hazards under the OSH Act are very
different than those used to evaluate unreasonable risks
under TSCA.

TSCA requires the assessment of risk to workers in the
absence of PPE, and if risks are identified, it can then
be considered whether the risks would or would not be
mitigated by PPE.

TSCA § 9(b) provides that EPA "shall coordinate
actions taken under [TSCA] with actions taken under
other Federal laws administered in whole or in part by
the Administrator" (15 U.S.C. § 2608(b)). While EPA
is supposed to coordinate the "actions" under each
statute, this provision does not contemplate EPA
excluding exposures from the risk analyses prepared
under TSCA. The remaining language of TSCA § 9(b)
highlights that Congress intended for EPA to prepare
risk evaluations analyzing all exposures, including
those that might be addressed under another authority.
The risk evaluation does not take into consideration
instances where PPE might NOT be used.

"[iIndividuals with impaired lung function due to
asthma, emphysema, or chronic obstructive pulmonary
disease ... may be physically unable to wear a
respirator."

Workers may choose to forego respirators because
they "may also present communication problems,
vision problems, worker fatigue, and reduced work
efficiency."	

of OSHA's PELs were issued shortly after adoption of the
Occupational Safety and Health (OSH) Act in 1970 and
have not been updated since that time." OSHA provides
an annotated list of PELs on its website, including
alternate exposure levels. For 1,4-dioxane, the alternates
provided are the California OSHA PEL of 0.28 ppm and
the ACGM TLV of 20 ppm.

(https://www.osha.gov/dsg/annotated-pels/tablez-l.html)
EPA's approach for developing exposure assessments for
workers and ONUs is to use the reasonably available
information and expert judgment. When appropriate, in
the risk evaluation, EPA will use exposure scenarios both
with and without engineering controls and/or PPE that
may be applicable to particular worker tasks on a case-
specific basis for a given chemical. Thus, while EPA has
evaluated worker risk with and without PPE, as a matter
of policy, EPA does not believe it should assume that
workers are unprotected by PPE where such PPE might be
necessary to meet federal regulations, unless it has
evidence that workers are unprotected. For the purposes of
determining whether or not a condition of use presents
unreasonable risks, EPA incorporates assumptions
regarding PPE use based on information and judgment
underlying the exposure scenarios. These assumptions are
described in the unreasonable risk determination for each
condition of use, in section 5.2. Additionally, in
consideration of the uncertainties and variabilities in PPE
usage, EPA uses the high-end exposure value when
making its unreasonable risk determination in order to
address those uncertainties. EPA has also outlined its
PPE assumptions in section 5.1. Further, in the final risk
evaluation for 1,4-dioxane, EPA has determined that most
conditions of use pose an unreasonable risk to workers

Page 89 of 212

-------


• On p. 135, EPA claims that an APF=10 respirator is
sufficient to eliminate even high-end inhalation non-
cancer risk "during industrial use." This is not
accurate: EPA found that an APF=25 respirator is
necessary to get the acute high-end MOE above the
benchmark MOE (Table 5-4) and that even an APF=50
respirator is not sufficient to get the chronic high-end
MOE above the benchmark MOE (Table 5-5).

even with the assumed PPE.

•	EPA did assess the risk to workers in the absence of PPE
and with PPE; are in Tables 4-4 through 4-11 in Section
4, Risk Characterization.

•	EPA has corrected the typographical error that appeared
on page 135 of the draft risk evaluation.

24,
30,
47,
52, 58

Other Conditions of Use

•	1,4-dioxane has been reported more than 500 times as
an ingredient (not as an impurity or byproduct) used
in hydraulic fracturing fluids across several states.
EPA failed to identify this as a condition of use.
Worker exposure to hydraulic fracturing fluids and
wastewater must be evaluated.

•	EPA did include spray foam; however, the
manufacturer claims that it's presence should be
considered an impurity or a byproduct and therefore
recommends EPA remove spray foam as a category in
the risk evaluation.

o EPA lacks monitoring data on spray foam use.
o Regarding spray foam use, it was also claimed
that EPA: 1) does not consider use of PPE in
its exposure scenarios, 2) overestimates the
number of workers on a per job basis, and 3)
does not account for ventilation activities, or
re-entry and re-occupancy times,
o EPA should work with the SPF industry to
develop more appropriate exposure estimates
for potentially exposed workers.

•	1,4-Dioxane may be produced as a reaction
byproduct, particularly in chemicals which are
produced by ethoxylation. EPA should evaluate those

• EPA disagrees with the commenter. While 1,4-dioxane
has been reported many times to FracFocus, the national
hydraulic fracturing chemical registry, the reported
concentrations are very low, and the reports include
presence of 1,4-dioxane with chemicals like ethoxylated
alcohols. The reported 1,4-dioxane is present as an
impurity or in the produced water. EPA initially excluded
production of 1,4-dioxane as a by-product from certain
other chemicals and presence as a contaminant in
industrial, commercial and consumer products from the
scope of the risk evaluation using EPA's discretion under
TSCA section 6(b)(4)(D). While EPA has addressed
some conditions of use related to 1,4-dioxane as a
byproduct in this risk evaluation, EPA expects that 1,4-
dioxane exposures associated with the use of ethoxylated
alcohols used in hydraulic fracturing fluids would be
considered in the scope of a risk evaluation for
ethoxylated alcohols. In cases like this, EPA believes its
regulatory tools under TSCA section 6(a) are better
suited to addressing any unreasonable risks that might
arise from these activities through regulation of the
activities that generate 1,4-dioxane as an impurity or
cause it to be present as a contaminant than they are to
addressing them through direct regulation of 1,4-dioxane.
This case-by-case approach for byproducts exposures is

Page 90 of 212

-------
COU in this risk evaluation, not in the future risk
evaluations for the ethoxylated chemicals.

Page 91 of 212

consistent with the various scenarios explained in the
Risk Evaluation Rule, 82 FR at 33730.

The commenter appears to be referring to submitted
comments from Johns Manville, a
distributor/manufacturer of spray polyurethane foam
(SPF). EPA acknowledges that Johns Manville claims
that the 1,4-dioxane's presence is as an impurity or
byproduct. EPA has also noted that there are many other
manufacturers of SPF products and patented technologies
exist. EPA has found data from other manufacturers that
suggests 1,4-dioxane might not always be a byproduct or
impurity. Due to contradicting and limited information
and the potential for worker exposure, EPA maintains its
assessment of SPF use as an occupational exposure
scenario.

In the final risk evaluation, eight consumer conditions of
use are evaluated based on the uses identified in EPA's
2015 TSCA Work Plan Chemical Problem Formulation
and Initial Assessment of 1,4-Dioxane (	:015).

An additional systematic review effort was undertaken
for consumer exposures to identify, screen, and evaluate
relevant data sources. These conditions of use include use
of 1,4-dioxane as a surface cleaner, antifreeze, dish soap,
dishwasher detergent, laundry detergent, paint and floor
lacquer, textile dye, and spray polyurethane foam (SPF).
1,4-Dioxane may be found in these products at low levels
(0.0009 to 0.02%) based on its presence as a byproduct of
other formulation ingredients {i.e., ethoxylated
chemicals). Inhalation exposures are estimated for
consumers and bystanders and dermal exposures are
estimated for users. Acute exposures are presented for all
consumer conditions of use, while chronic exposures are
presented for the conditions of use that are reasonably

-------




expected to involve daily use intervals {i.e., surface
cleaner, dish soap, dishwasher detergent, and laundry
detergent). See Section 2.4.3 of the final risk evaluation.

22,
24,
48,
58, 59

Occupational Dermal Exposure

•	EPA needs to clarify what absorption values and
assumptions were applied to each scenario for
estimating dermal exposures.

o Risk values for the scenarios without gloves are
reduced exactly by the protection factor (PF)
EPA assumed for the three scenarios with
gloves. See Table 5-10 on p. 144 and Table 5-
11 on p. 145. This should not be the case if
EPA applied different values for skin
absorption for the scenarios with and without
gloves.

o It is not clearly stated why EPA did not choose
the more conservative absorption rate (3.2%)
for non-occluded/no glove scenarios supported
inMarzulli etal., (1981).
o EPA reports that there are numerous ways in
which glove use could increase skin exposure
through occlusion; however, it is not clear how
EPA's analysis accounted for this,
o It is not clear why EPA cited risk estimates for
the 20x PFs in the risk determination section
but ignored higher risks found with no gloves
or gloves with lower PFs.

•	EPA needs to provide a more thorough analysis
surrounding the limitations and protections offered by
glove use and should also account for more recent
data, such as Dennerlein et al., (2013).

•	Tables 5-4, 5-5, 5-9, 5-10, and 5-11 do not distinguish
between high-end and central tendency exposures.

•	The absorption values without gloves are discussed in
Section 2.4.1.14 (Dermal Exposure Assessment).

Materials of constructions of gloves and their suitability
for 1,4-dioxane are described under sub-section Dermal
Exposure Estimation. Since the absorption and penetration
data for 1,4-dioxane through various materials of gloves
are not available, the protection factor was used in
computation of risk values.

•	As indicated in this response to comments document, the
dermal exposure calculations have been updated and
validated by using sensitivity analysis and by comparing
test results reported by the researchers at the Kansas State
University, Manhattan, Kansas; and at the University
Erlangen-Nuremberg, Erlangen, Germany. The above-
mentioned references are cited in the revised risk
evaluation document (see Section 2.4.1.14 - Dermal
Exposure Assessment). The revised risk evaluation
document has been updated by deleting Bronaugh (1982)
and the relevant paragraph as the Bronaugh (1982) cited
data are not used in the dermal exposure assessment. The
dermal calculations have been updated in the revised risk
evaluation by including a conceptual diagram, tiered
analysis using: a) updated calculations; b) sensitivity
analysis to evaluate chemical fluxes at various fractional
absorption varying from negligible (~0) to complete
absorption (1.0); c) overall comparison of modeled
chemical fluxes with in-vitro and ex-vivo test data
reported in the literature.

•	EPA included a thorough analysis on incorporation of
glove protection, limitations and protections of glove use,

Page 92 of 212

-------
Some cancer risks exceed benchmarks even with
respirator/glove use.

Although assumptions regarding glove use were
clearly stated for dermal calculations, the empirical
basis for the protection factors needs empirical
justification. For example, the "fraction absorbed" into
the stratum corneum was based on a "large dose"
assumption not verified in the analysis.

EPA must explain why it relies on the in vitro dermal
absorption study by Bronaugh (1982), which is a
secondary source book chapter, to estimate dermal
absorption.

o This study is not publicly available and has not
been subject to any quality review, yet
information from this source is used in
calculating (HEDs), which are themselves used
as a basis for reducing the interspecies
uncertainty value from 10 to 3 (pp. Ill, 118).
EPA cites the fractional absorption potential for 1,4-
dioxane to be 0.86 or 0.78, depending on the setting,
based on Kasting and Miller (2006). EPA then adjusts
the 0.86 or 0.78 values by the 0.3% or 3.2% values
based on Bronaugh (1982). Through this "double"
adjustment, potential dermal risk is underestimated.
EPA also relies on Marzulli etal., (1981), which
examined absorption in adult rhesus monkeys. But the
vehicles employed were methanol and skin lotion, and
it is not clear how representative they are of absorption
under the conditions of use in this risk evaluation.
Moreover, the authors describe their results as
providing only "crude estimates."

EPA failed to review a dermal absorption study
conducted by Dennerlein et al., (2013). This study

potential for occlusion in Appendices E (E5 and E6), and
G. Additional information are also included in
Supplemental document (Engineering Assessment of
Occupational Exposure for 1,4-Dioxane).

Analysis and interpretations have been updated in the
revised risk evaluation document including citations of
Dennerlein et al., and other recent publications (see
Section 2.4.1.14 for details on the multiple citations).
EPA updated dermal tables 5-9, 5-10, and 5-11 to include
central tendency and clearly delineate between high-end
and central tendency exposures.

A sensitivity analysis has been performed with respect to
fraction sorbed. The double adjustment is no longer
applicable. The citation of Bronaugh (1982) was included
to provide general description on dermal absorption. EPA
no longer relies on this source as the basis for any
quantitative analysis. The revised risk evaluation document
has been updated (to be done) by deleting the relevant
paragraph as these data are not used in the revised
assessment. In the final risk evaluation, Bronaugh (1982) is
briefly referenced in the discussion of toxicokinetics to
provide a more complete picture of the reasonably available
information on dermal absorption, but data from the
Bronaugh (1982) is not used in subsequent analysis. While
EPA is unable to post the copyrighted material, the book is
publicly available.

The dermal calculations have been updated by including a
conceptual diagram, tiered analysis using: a) updated
calculations; b) sensitivity analysis to evaluate chemical
fluxes at various fractional absorption varying from
negligible (~0) to complete absorption (1.0); c) overall
comparison of modeled chemical fluxes with in-vitro and
ex-vivo test data reported in the literature. There is no

Page 93 of 212

-------
could not be found in either the Risk Evaluation or the
Systematic Review Supplemental File and should be
made available for public review.

•	Some of the assumptions applied in the dermal
exposure models were not well supported by the
weight of scientific evidence. It is suggested that EPA
revisit these models and ensure the assumptions
properly reflect occupational working conditions.

o No justification was provided for the

assumption that dermal exposures occur as one
event/day.

•	Contradictory assumptions about the absorption of a
"large dose" were identified:

o In one section, it was assumed "the chemical
amount forms a residual layer (or pool) on the
top of the skin... [that] acts as a reservoir to
replenish top layers [of skin] as the chemical
permeates into the lower layer."

o Elsewhere, also regarding "large doses", it was
stated "Only a fraction of the 1,4-dioxane that
contacts the skin will be absorbed as the
chemical readily evaporates from the skin,"
given its volatile nature. It is not clear whether
the evaporated fraction was considered in the
inhalation exposure.

double adjustment in the updated calculations in the
revised risk assessment document.

Marzulli et al., (1981) and Bronaugh (1982) were only
toxicological section. The revised document has been
updated by deleting the relevant paragraph for dermal
exposure as these data are not used in the assessment.
The Dennerlein et al., (2013) reference has undergone
systematic review and the results can be found in the
Systematic Review Supplemental File. The source can be
found in HERO (HERO ID: 3537857).

A section describing the general approaches and methods
for dermal exposures of 1,4-dioxane from the outer
surface of the skin to the inner layers of the skin are
included in the revised document (see Section 2.4.1.14 -
Dermal Exposure Assessment).

A search of the document text shows that the commenter
is referring to a passage in section 3.4.1.14 and a passage
in Appendix G.7.1. The comment noting that
contradictory assumptions exist is inaccurate has been
clarified and the statements in the revised risk evaluation
document are accurate.

Section 3.4.1.14 is a summary that is dermal contact with
1,4-dioxane and despite the commenter's statements,
never refers to a "large dose" or similar concept. This
section is talking about dermal exposure in a general
sense, noting that because of 1,4-dioxane's volatility, not
all of the applied chemical will be absorbed into the skin.

The passage in Appendix G.7.1 is referring to a section
describing the mathematical basis for the dermal model. In
this section, titled Large Doses (Case 2: MO > Msat), it is
clearly describing a "large dose" as one where the initial

Page 94 of 212

-------




dose is greater than that required to saturate the upper
layers of the stratum corneum. This is the only section
where the term "large dose" appears in the risk evaluation.
The passage goes on to say that "in this case, absorption
and evaporation approach steady-state values as the dose
is increased...", showing agreement with the general
statement presented in section 3.4.1.14.

58

Aggregate Exposures

•	EPA should consider total daily intake by combining
exposures from inhalation and dermal, as well as oral,
routes.

o Worker exposure and risk are understated
because risk levels for dermal and
inhalation exposure were determined
separately.

¦ On the other hand, it was noted that
for inhalation exposure calculations,
none of the exposure estimates
considered active ventilation
controls which could yield an
overestimate,
o EPA assumed there are no oral exposures to
workers, however, hand-to-mouth exposure
is likely common.

•	Combined inhalation and dermal exposures were also
not considered (review Take etal., 2012).

•	EPA is required to evaluate exposures from
combinations of activities. Exposure to workers in
occupational settings should be added to exposure of
those workers in non-occupational settings.

• TSCA section 6(b)(4)(F)(ii) directs EPA to "describe
whether aggregate or sentinel exposures to a chemical
substance under the conditions of use were considered,
and the basis for that consideration" in risk evaluations.
EPA defines aggregate exposures as the combined
exposures to an individual from a single chemical
substance across multiple routes {i.e., dermal, inhalation,
or oral) and across multiple pathways {i.e., exposure from
different sources). 40 CFR 702.33. EPA defines sentinel
exposures as the exposure from a single chemical
substance that represents the plausible upper bound of
exposure relative to all other exposures within a broad
category of similar or related exposures. 40 CFR 702.33.
EPA considered the reasonably available information and
used the best available science to determine whether to
consider aggregate or sentinel exposures for a particular
chemical. EPA has determined that using the high-end
risk estimate for inhalation and dermal risks separately as
the basis for the unreasonable risk determination is a best
available science approach. There is low confidence in the
result of aggregating the dermal and inhalation risks for
this chemical if EPA uses an additive approach, due to the
uncertainty in the data. EPA does not have data that could
be reliably modeled for the aggregate exposure, such as
would occur with a PBPK model. Using an additive
approach to aggregate risk in this case could result in an

Page 95 of 212

-------




overestimate of risk. Given all the limitations that exist
with the data, EPA's approach is the best available
science. EPA has added language to the Key
Assumptions and Uncertainties section describing these
assumptions and uncertainties.

24,
46,
48, 58

Other comments related to worker exposure

•	EPA's risk determinations for workers emphasize
central tendency exposure levels and give less weight
to high-end exposures. How is the central tendency
scenario more protective of workers?

•	EPA, for each category of worker, must identify and
evaluate the worker whose exposure represents the
plausible upper bound of exposure (sentinel exposure).

• EPA defines "sentinel exposure" to "mean the
exposure from a single chemical substance that
represents the plausible upper bound of
exposure relative to all other exposures within
a broad category of similar or related
exposures."

• This definition was not applied
throughout the risk evaluation.

•	Baseline exposure to workers via consumption of
drinking water and product use, as well as
environmental releases to air, water, and land, should
be incorporated into worker risk calculations.

•	The agency assumes that all workers are "healthy" in
its risk characterization (p. 132), but also
acknowledges that there may be numerous worker
subpopulations that may have pre-existing conditions
that affect the liver or other organs targeted by 1,4-
dioxane (p. 108). These sensitive subpopulations are
not included in its analysis of risk.

•	In the Monte Carlo analyses to determine percentile of

•	EPA examines the totality of risk estimates for a condition
of use when making a determination of unreasonable risk.
EPA makes one determination for each condition of use
and describes the basis in terms of risks to workers and
ONUs. In the risk evaluation for 1,4-dioxane, EPA used
the high-end exposure value when considering worker
risks in order to address the uncertainties and variability in
PPE usage. For inhalation exposures, EPA, where
possible, estimated ONU exposures and described the
risks separately from workers directly exposed. To
account for those instances where, based on EPA's
analysis, the monitoring data or modeling data for worker
and ONU inhalation exposure could not be distinguished,
EPA considered the central tendency risk estimate when
determining ONU risk. EPA considered sentinel exposure
the highest exposure given the details of the conditions of
use and the potential exposure scenarios. Sentinel
exposures for workers are the high-end no PPE scenario
within each OES. In presenting the inhalation results, high
intensity use was characterized by the model iteration that
utilized the 95th percentile duration of use and mass of
product used and the maximum weight fraction derived
from product specific SDS, when available. Dermal
exposures for workers for high intensity use were
characterized by the model iteration that utilized the 95th
percentile duration of use and maximum weight fraction.

•	Please refer to section 1.4.2 in the risk evaluation which
provides details as to why exposures related to drinking

Page 96 of 212

-------
exposure distribution, EPA should select parameters
based on an assessment of present-day conditions, with
a clear description of the underlying assumptions and
present-day information reviewed.

•	The higher upper bound for the saturation
factor parameter is not sufficiently justified.

•	The assigned parameters are outdated and
should be re-evaluated.

•	EPA lumps together a highly diverse set of uses as
"industrial uses" (p. 58) and asserts that all such
operations "are expected" to be similar. Support for
this assumption should be provided.

•	EPA uses single scenarios to represent each of
the following activities despite their varied
nature: all processing scenarios other than
repackaging (p. 163); all intermediate use
scenarios (p. 165); all open system functional
fluid use scenarios (p. 166); all laboratory
chemicals use scenarios (p. 168); and all
disposal scenarios (p. 175).

•	No data or analysis was provided to show these
scenarios are representative of other scenarios
within a grouping or that this provides a health-
protective approach.

• EPA should provide support for the
assumption that exposure via the
selected scenario is representative of all
potential worker scenarios, and that the
selected exposure scenario was the most
significant.

•	EPA assumed that occupational non-users (ONUs)
would have lower exposure levels because they do not
typically directly handle the chemical. No exposure

water as well as other exposure pathways related to
environmental releases were not included in the risk
evaluation.

For worker exposures, EPA does account for exposures to
potentially exposed or susceptible subpopulations (PESS)
by using the high-end exposure value when making its
unreasonable risk determination for workers.

The model parameter values used represent EPA's current
state of knowledge of the various parameters (e.g.,
saturation factor, ventilation rate, etc). All parameters are
cited and discussed in the Appendix G.

EPA has categorized the different industrial users based
on where sufficient process information is known. While
EPA recognizes the limitations of grouping the conditions
of use enumerated in this comment, the current grouping
is sufficient to represent general exposure within each use
category.

EPA recognizes that sufficient data does not exist for a
more quantitative assessment of ONU exposures. The risk
evaluation takes known process descriptions, facility
designs, and other factors into account when assessing the
exposure potential for ONUs. The risk evaluation and
supplemental documents note that while ONU inhalation
exposures are assumed to be lower than inhalation
exposures for workers directly handling the chemical
substance, in some cases ONUs may have higher
exposures than workers when workers wear PPE and the
ONUs do not. Further, EPA has also used a different
approach in the final 1,4-dioxane risk evaluation for
making unreasonable risk determinations on ONU

Page 97 of 212

-------


data for ONUs were evaluated, and it was noted that
this assumption is likely only valid if the ONUs were
also wearing PPE.

• ONU exposures were not quantitatively
assessed due to "lack of data".

•	The risk level for the entire worker population is set at
the same level that EPA set for the most exposed
individual in a population. This resulted in the
determination that the majority of conditions of use
posed no risk to workers. This approach must be
rejected on scientific, as well as legal, grounds.

•	The range of workers that EPA defines as ONUs is too
large to support a single classification. Supervisors and
managers have very different exposure patterns than
skilled trade workers, yet all three are assumed to face
similar risks under EPA's ONU categorization.

exposures, i.e., looking to the central tendency risk for
workers where EPA is unable to separately calculate
worker and ONU risks.

•	As noted in the draft risk evaluation, EPA relied on
Agency precedent and NIOSH guidance when choosing
the 10"4 cancer risk benchmark to evaluate risks to
workers from 1,4-dioxane exposure.

•	The standard cancer benchmarks used by EPA and other
regulatory agencies range from 1 in 1,000,000 to 1 in
10,000 (i.e., lxlO"6 to lxlO"4) depending on the
subpopulation exposed.

•	EPA, consistent with 2017 NIOSH guidance, used lxlO"4
as the benchmark for the purposes of this unreasonable
risk determination for individuals in industrial and
commercial work environments, including workers and
ONUs lxlO"4 is not a bright line and EPA has discretion
to make unreasonable risk determinations based on other
benchmarks as appropriate. See section 5.1.1.2 of the risk
evaluation for additional information.

•	EPA recognizes the commenter's request for further
granulation of the ONU category but also recognizes that
every workplace is different with an array of worker types
experiencing varying types and degrees of exposures
based on site design, equipment, company culture, etc.
Without reasonably available information about each
facility EPA cannot produce further divisions of ONU
designations.

48,
58, 59

Modeling

•	EPA should verify model outputs by using a tiered
approach towards exposure to ensure model outputs
represent exposure levels in line with real-world
conditions.

•	Dermal modeling with ChemSTEER could be

•	Tiered approach has been considered using published test
results and sensitivity analysis in the revised version.

•	The revised risk evaluation document has been updated
with inclusion of several dermal exposure scenarios of
1,4-dioxane from the IHSkinPerm© (developed by
American Industrial Hygiene Association, AIHA) output

Page 98 of 212

-------
validated with other, higher-tier, modeling programs,
such as IHSkinPerm. This method would be consistent
with techniques discussed in the trichloroethylene
problem formulation. A tiered approach to exposure
modeling would be helpful to triangulate accurate risk
levels and confirm underlying assumptions, such as the
paragraph on p. 76 that states that dermal exposure
with gloves will have higher measured absorption
values, despite 1,4-dioxane's highly volatile nature.

It is suggested that EPA add a sensitivity analysis to
identify the key factor(s) to pinpoint overly
conservative assumptions.

A clearer description about the assumptions contained
within each model is required. Nested assumptions and
uncertainties can lead the models to provide unrealistic
exposure levels. This is particularly true in the dermal
exposure calculations, where model inputs are not
supported by the weight of scientific evidence.
Regarding modeling, it is suggested that EPA provide
added clarity when modeling exposures by organizing
assumptions into a table and discussing the rationale
for each assumption in the corresponding modeling
section.

For inhalation exposure models, the lower ventilation
rate assumption used in the models was created in the
1980s and does not reflect modern day design
standards and facilities. Thus, it is impossible to
conclude with confidence that the high-end exposure
predicted by the EPA model is less than the 99.9th
percentile.

Factors viewed as modeling errors:

o Extrapolation from inhalation to dermal risks
	without considering flux dynamics that are

are summarized in Section 2.4.1.1.12. Description of
conceptual diagram, synopsis of existing tools/models,
interpretations, and citations of references are also
included in the revised risk evaluation document.

•	Sensitivity analysis and evaluations with respect to
chemical flux have been included in the revised version of
risk evaluation document.

•	EPA recognizes the need for a clear and simple way to
understand the assumptions made within each model and
the associated rationale. The approach, assumptions, and
mathematical calculations are addressed in Appendix G.

•	A Monte Carlo simulation was performed using wide
variations of recent engineering published data to cover
applicable ranges for saturation factor, ventilation rate,
mixing factor and others. There is no additional
ventilation (e.g., fan) modeled in this scenario.

•	Revisions have been made by introducing a conceptual
diagram, general approaches and methods for dermal
exposures of 1,4-dioxane from the outer surface of the
skin to the inner layers of the skin, tiered analysis using a
sensitivity analysis and comparing calculated chemical
fluxes from the model used and dermal test data from in
vitro and ex vivo studies (see Section 2.4.1.14 - Dermal
Exposure Assessment). A diagram in the revised risk
evaluation document (Figure 2-2) shows the comparative
transdermal flux parameters for 1,4-dioxane across human
skin at various exposure conditions.

Page 99 of 212

-------


uniquely applicable to dermal absorption,
o Assumption that time equals infinity in dermal
modeling (which overestimates evaporation
and underestimates absorption),
o Crucial mistakes in dermal dose equation

calculations (not further elaborated).

- TTsc of fixed dove PFs of 5\. 1 Ox. and 2 fix



Kxposiirc-UclaU'ri Systematic Review

24,
58, 59

Inappropriate Scoring

•	A highly relevant OSHA monitoring study was
inappropriately excluded because the data were in a
text file instead of an Excel file (see p. 105 of the 1,4-
dioxane Supplemental File: Data Quality Evaluation of
Environmental Releases and Occupational Exposure
Data). The study was scored as unacceptable because
[the data are] "all smooshed together in a text file and
not useful." This is arbitrary and illustrates inconsistent
application of scoring criteria; file format is not one of
the criteria for scoring a study as unacceptable. EPA
should have worked with OSHA to obtain the data.

•	2016 BASF Data: Why did EPA assign a score of 1 to
"Sample Size" and include a note indicating
"Representative sample size," when the data set
comprised only 28 samples from a single site? In the
Risk Evaluation itself, EPA acknowledges that these
data are unlikely to be representative: "It is uncertain
to what extent the limited monitoring data used to
estimate inhalation exposures for this scenario that
could be representative of occupational exposures in
other manufacturing facilities of 1,4-dioxane" (p. 55).
Sampling rate was missing in some cases, so it was
assumed that reported rates applied to all measures.

•	EPA converted the document into a file format that was
intelligible and re-examined the data. The data is OSHA
sampling data from 2016 and all of the entries are non-
detects. The scoring for the document was upgraded from
unacceptable to medium. Per OSHA recommendation,
however, EPA policy is to not use non-detects from
OSHA because neither OSHA nor EPA can determine if
the samples were collected from an environment actually
containing 1,4-dioxane. Using these samples may
underestimate occupational exposure. Given that all of the
data points were non-detects for 1,4-dioxane, the current
monitoring data used in the risk evaluation does not
change.

•	Document scores are determined before the risk
evaluation and in isolation relative to other sources, in
other words scores are purely based on the merits of the
document itself. The scoring rubric in the Application of
Systematic Review in TSCA Risk Evaluations provides the
following description for a score of 1 in the Sample Size
category; "Statistical distribution of samples is fully
characterized." The 2016 BASF data is only from one site,
but the statistical distribution of samples from that site are
fully characterized and representative of the site.
Furthermore, because there are only two known
manufacturers or importers in the USA, the data is

Page 100 of 212

-------
• 2017 BASF Data: Why did EPA assign a score of 2 to
"Sample Size," when the data set comprised only four
data points from a single site?

representative of the entire industry.

The scoring metrics for "Samples Size" in the monitoring
data scoring rubric of the Application of Systematic
Review in TSCA Risk Evaluations is defined by how well
a range of samples are presented, not the actual number of
samples. The title of the metric is misleading in this way,
but the underlying definitions are clear. A document with
a score of 2 if defined as the "distribution of samples is
characterized by a range with uncertain statistics".
Furthermore, when scoring documents, all metrics are
based solely on the quality of the data within the
document. In this case, the 4 data points fall into the
Medium (score = 2) category.	

Page 101 of 212

-------
5. Human Health

Charge Question 5.1: Please comment on the evaluation of human health hazards including evaluation of portal of entry and
systemic toxicity for cancer. Are there any additional 1,4-dioxane specific data and/or information that should be considered?

Charge Question 5.2: Please comment on the evaluation of human health hazards including evaluation of portal of entry and
systemic toxicity for non-cancer. Are there any additional 1,4-dioxane specific data and/or information that should be considered?

Charge Question 5.3: Please comment on any other aspects of the human health risk characterization that have not been mentioned.

Charge Question 5.4: Please comment on the mode of action discussion and provide feedback on mode of action analysis.

Charge Question 5.5: Please provide comment on the presented approaches. Please provide comment on any additional model
consideration that EPA could include for cancer characterization.

Summary of Comments lor Specific
Issues Related to Charge Question 5

KPA/OPPT Response

SACC,

22, 58,
59

(jenotoxicity- Narrative and Weight of Kvidence

SACC Recommendation: Clarify the reasoning leading
to the weight of evidence statement on page 96.
Immediately after discussing a study in which a
significant increase in point mutations was seen, the
TSCA document states: "Therefore, EPA concluded that
the weight-of-the- scientific evidence supports that 1,4-
Dioxane is not mutagenic but may elicit clastogenicity in
vivo at high doses." The study does provide some
evidence that 1-4- Dioxane is mutagenic. However,
because the evidence for the induction of gene mutations
in vivo comes from a single dose in one experiment, the
Committee member cautions about drawing a positive
conclusion about a mutagenic mode of action from the Gi
etal., (2018) study alone. The member recommended that
at this time it would be more scientific to state that there
is insufficient evidence to conclude that 1,4-Dioxane is

To belkT support wcighl of c\ idcncc conclusions ivlalcd lo
mutagenicity and genotoxicity, EPA performed data quality
review for all studies considered in this section. EPA revised
the narrative and associated appendix table of all genotoxicity
studies to reflect this change. EPA also made minor edits to
improve clarity of the narrative in Section 4.2.3.2.

For 1,4-dioxane, EPA concluded that there is insufficient
evidence to determine whether 1,4-dioxane or its metabolites
act through a mutagenic or otherwise genotoxic mode of
action. EPA also reviewed evidence for several plausible
MO As and concluded that there is insufficient evidence to
determine the mechanism of action for carcinogenicity of
1,4-dioxane for any of the tumor locations. While some
evidence for the proposed MOA of metabolic saturation
followed by cytotoxicity and regenerative proliferation is
available for liver tumors, the available evidence is also
consistent with alternate plausible MO As (as outlined in

Page 102 of 212

-------
mutagenic or induces cancer through a mutagenic mode
of action.

General Public Comment:

•	The analysis is performed in a narrative format that is
difficult to follow and, in many places, appears
contradictory.

Public Comments in support of a non-genotoxic MOA

•	In the draft risk evaluation, EPA agreed with the IRIS
assessment conclusion that 1,4-dioxane is not
genotoxic in the absence of cytotoxicity but then
incorrectly interpreted additional data as a positive
indication of genotoxicity.

•	Data from Goldsworthy el al., (1991) indicates that
the observed stimulation of DNA synthesis at elevated
exposure levels did not represent a genotoxic event,
but rather is a key event for regenerative cell
proliferation and/or tumor promotion.

•	Studies by Itoh and Hattori (2019), Gi et al., (2018),
and Morita and Hayashi (1998) should be re-evaluated
as the study authors argue the observed effects were
due to non-genotoxic mechanisms.

•	Observations of some genotoxicity at cytotoxic
extremely high levels is not relevant to a genotoxic
MOA at lower doses.

Public Comments in support of a genotoxic MOA:

EPA dismisses key data relevant to genotoxicity and
independent scientific conclusions in support of genotoxic
mechanisms.

•	Gi et al., (2018) concludes that "1,4-dioxane is a	

Page 103 of 212

Appendix I). Applying a threshold approach to evaluate
cancer risk for 1,4-dioxane would not be adequately
supported by available mechanistic evidence. Consistent with
EPA guidance, EPA performed BMD analysis on tumor data
and applied the best fit models for the data

In response to SACC and public comment, EPA has slightly
modified the weight of evidence conclusions of the
genotoxicity narrative to be more precise. The final risk
evaluation now concludes the section on genotoxicity as
follows: "Based on the weight of scientific evidence, EPA
concluded that there is some evidence for genotoxicity in
vivo at high doses, but there is insufficient evidence to
conclude that 1,4-dioxane is mutagenic or induces cancer
through a mutagenic mode of action." EPA arrived at this
conclusion based on the weight of scientific evidence with an
awareness of the potential for differences in genotoxicity
across endpoints and tissue types.

-------


genotoxic hepatocarcinogen and induces
hepatocarcinogenesis through a mutagenic MOA."
However, EPA states "the weight of scientific
evidence supports that 1,4-dioxane is not mutagenic"
(p. 96). The "weight of evidence" approach that EPA
utilized to reach this conclusion should be clearly
outlined.



24

Genotoxicity- Historical controls

Public Comment: Itoh and Hattori (2019) discounted the
statistically significant increase in micronucleated
immature erythrocytes (MNIE) because these changes
were within the historical control range. EPA cancer
guidelines state that statistically significant increases in
tumors should not be discounted simply because
incidence rates in the treated groups are within the range
of historical controls

The equivocal findings for MNIE in rats reported by Itoh and
Hattori were consistent with the mix of both positive and
negative studies for this endpoint reported in other studies.
The weight of evidence for this endpoint supports EPA's
overall conclusion that there is some evidence for
genotoxicity in vivo at high doses. The EPA cancer
guidelines refer specifically to tumor incidence and are not
directly applicable to micronucleus assays.

58

Genotoxicity- Studies to consider

Public Comment: The following list of studies are in a
support of a genotoxic MOA: meiotic nondisjunction
(Munoz and Barnett, 2002); micronucleus formation (i.e.
clastogenic activity) (Mirkova, 1994; Morita and Hayashi,
1998; Roy et al., 2005; Itoh and Hattori, 2019); point
mutations (Gi et al., 2018); single-strand breaks (Sina et
al., 1983; Kitchin and Brown, 1990); and replicative DNA
synthesis (Miyagawa et al., 1999).

All of these studies were considered and contributed to the
weight of evidence for genotoxicity. EPA evaluated data
quality for each of these studies.

22

Genotoxicity- ToxCast Screening Assay Evidence

Public Comment: Regarding ToxCast results, EPA
incorrectly states that 1,4-dioxane was observed to

Data for the assay EPA describes can be found on the EPA
Chemistry Dashboard bv searching for 1,4-dioxane and
clicking on the ToxCast summary. The relevant assay names
are TOX21 p53 BLA p4 ch2 and

Page 104 of 212

-------


increase the transcriptional activity of the p53 tumor
suppressor protein in human colon cancer cells (HCT116)
24 hours after 1,4-dioxane exposure, and that this is
indicative of an active DNA damage and repair response.
However, the ToxCast results for 113 assays on 1,4-
dioxane were all negative and do not support a conclusion
that there is any activity associated with DNA damage
and repair. It is unclear how EPA weighed the evidence
from these ToxCast studies and integrated the disparate
results (compared to IRIS) to arrive at a conclusion.

TOX21_p53 BLA_p4 ratio

Mode o

' Action Com incuts

SACC

MOA - Apply the MOA Framework

SACC Recommendation: The EPA needs to explain and
follow its guidelines for evaluating the MOA Framework.

SACC Recommendation: Clarify and expand on
environmental exposure and MOA.

EPA has substantially modified the discussion of plausible
MO As for 1,4-dioxane carcinogenicity. EPA now more
explicitly describes and follows the MOA framework to
evaluate evidence for a specific proposed MOA in the new
Appendix I.

SACC,
59

MOA- Role of Metabolic Saturation

SACC Recommendation: The section discussing the
importance of 1,4-Dioxane metabolism in its MOA needs
to be edited. It should be made clear that high systemic
concentrations of 1,4-Dioxane does not necessarily
indicate metabolic saturation but could result from
decreasing hepatic blood flow. And, if there is extensive
first pass clearance in the liver, then overall hepatic
metabolic clearance may be perfusion limited.

Public Comment: The draft risk evaluation concludes
that the data from Kasai et al., (2009) do not support
saturation from inhalation exposures since the blood
levels of 1,4-dioxane increased linearly with doses above

EPA has included a new appendix applying the MOA
framework to evaluate the hypothesis that the MOA of 1,4-
dioxane carcinogenicity is related to metabolic saturation. In
general, EPA concluded that there is insufficient evidence to
conclude that metabolic saturation is a necessary key event in
liver carcinogenesis.

EPA considered the alternate hypothesis presented by the
SACC member that reduced rates of metabolism may be due
to reduced liver perfusion at high doses. EPA concluded that
it is unlikely for liver toxicity from the single dose
administered in the toxicokinetic study to lead so rapidly to
reduced metabolism due to reduced liver function.

Kasai et al., 2008 report that plasma levels of 1,4-dioxane

Page 105 of 212

-------


400 ppm (-400 mg/kg/day). However, the information
presented by Kasai et al., is not sufficient to identify a
transition from the low dose, first-order metabolism and
the higher-dose, zero order threshold since the doses
exceed the reported threshold of 30-100 mg/kg/day in
rats. Although Kasai et al., does not provide information
for the lower doses used in the study (100 and 200 ppm),
the omission of this information suggests that 1,4-dioxane
blood levels were below the limit of detection -
suggesting that saturation occurred somewhere between
100 and 400 ppm.

were detected in the concentrations of exposure to 400 ppm
and above, and increased linearly with the increase in the
exposure concentration. These observations are consistent
with linear first order kinetics. Had saturation occurred
between 100 and 400, we would expect plasma
concentrations at higher levels of exposure to increase in a
non-linear fashion consistent with Michaelis-Menten kinetics.
The threshold for metabolic saturation proposed based on
evidence in oral exposure studies cannot be assumed to apply
to inhalation exposures because of differences in kinetics
across exposure routes (e.g. first-pass metabolism).

48, 22,

MO A- Proposed MOA for Liver Tumors

Public Comments:

•	The non-mutagenic, alternative threshold, high-dose-
induced cytoproliferative MOA was not carried
through the entire analysis and was not included in the
risk characterization.

•	The EPA is encouraged to further/more closely
consider the following studies which support the
threshold MOA: Dourson et al., 2014; Dourson etal.,
2017; Julien et al., 2009; Boobis et al., 2009; Young
etal., 1978; Nannelli et al., 2005; Young etal., 1977;
Sweeney et al., 2008; Woo et al., 1977; and US Army
Public Health Command, Studies on metabolism of
1,4-dioxane, Toxicology Report No. 87-XE-08WR-
09, Aberdeen Proving Ground, MD (March 2010);
and Toxicology and Environmental Research and
Consulting (TERC), Investigating the mode of action
for 1,4-dioxane-induced liver tumors in B6D2F1/Crl
mice, Midland, MI. (2019), Report to be submitted to
EPA.

In a new Appendix I, EPA methodically applies the MOA
framework outlined in the Guidelines for Carcinogen Risk
Assessment to evaluate evidence for the MOA for liver
tumors proposed by Dourson et al., (2014, 2017). In the
proposed MOA, metabolic saturation leads to accumulation
of the parent compound followed by cytotoxicity and
regenerative proliferation. This expanded analysis
incorporates more complete data summaries from Kociba
(1974) and the reanalyzed data from NCI (1971) reported in
McConnell (2013). The specific issues raised in these public
comments are addressed in that appendix.

In the risk evaluation, EPA has reviewed and discussed
Dourson et al., 2014, Dourson et al., 2017, Nannelli et al.,
2005, Young et al., 1998, Young etal., 1977; Sweeney et
al., 2008; Woo et al., 1977 and many other studies related to
metabolism and MOA. While some of these studies provide
evidence that is consistent with the proposed hypothesis,
evidence in these studies cannot rule out alternate
hypotheses. Furthermore, evidence in several studies
indicates that key events in the MOA proposed by Dourson et
al., are not necessary precursors to tumor formation, lending

Page 106 of 212

-------




to support to alternate hypotheses. Ultimately EPA concluded
that there is insufficient evidence to support a specific MOA
for liver carcinogenicity.

22

MOA- Specific evidence to consider in support of a
proliferative regenerative repair MOA

Public Comment: The EPA should re-evaluate and
consider the following data as support of a proliferative-
regenerative repair MOA:

•	The increase in hepatocellular foci in mid and high
dose males, and in high dose females in the study by
Kano etal., (2008). These occur at doses that exceed
the limit of metabolic saturation.

•	Kociba et al., (1974) reported hepatic degeneration
and regenerative hyperplasia at or below dose levels
that produced liver tumors, but incidence for these
effects was not reported.

•	EPA reported that nuclear enlargement in the
respiratory and olfactory epithelium in a chronic
exposure assay was not adverse; however, these could
also be key event changes observed with tumor
promotion.

As described above, EPA has added a new Appendix I that
methodically applies the MOA framework outlined in the
Guidelines for Carcinogen Risk Assessment to evaluate
evidence for the MOA for liver tumors proposed by Dourson
etal., (2014, 2017). In the proposed MOA, metabolic
saturation leads to accumulation of the parent compound
followed by cytotoxicity and regenerative proliferation.

EPA considered all of the specific evidence outlined in these
comments, including evidence of hepatocellular foci reported
in Kano (2008) and evidence of hyperplasia reported in
Kociba (1974). EPA weighed the available evidence,
including each of the arguments articulated in these
comments, and concluded that there is insufficient evidence
to support the proposed MOA. While the evidence
highlighted in these comments is consistent with the
proposed hypothesis, alternate hypotheses cannot be ruled
out. Other evidence suggests that not all of the proposed key
events are necessary precursors to tumor formation. For
example, evidence from Kano 2009 indicates increased
incidence of liver tumors at doses below those associated
with hepatocellular toxicity. This suggests that cytotoxicity
may not be a necessary key event in 1,4-dioxane exposure
leading to liver carcinogenesis. In addition, there are several
datagaps that prevent EPA from identifying causal
relationships between key events. For example, there is no
clear evidence demonstrating that cytotoxicity is a necessary
precursor to observed cell proliferation and there is not clear
evidence that 1,4-dioxane rather than a metabolite is the toxic
moiety. 1,4-dioxane or a metabolite may lead to cell

Page 107 of 212

-------




proliferation through an alternate mechanism.

59

MOA- Use of new ACC data

Public Comment: The ACC has sponsored a new 90-day
repeated dose study that was submitted along with
comments. This study demonstrates a clear threshold for
any effect in the liver at a genomic level. Results of this
study are consistent with those in the 13-week drinking
water study reported by Kano et al., (2008) in which
BDF1 mice were exposed to up to 25,000 ppm of 1,4-
dioxane in drinking water. Results are consistent with
changing metabolic competency of the female mice as a
critical key event in 1,4-dioxane toxicity. The available
evidence points to the parent 1,4-dioxane as the toxic
species. A direct mitogenic response is triggered in the
liver of female mice. This mitogenic response occurs
relatively early and likely adds to the regenerative repair
that is suggested with the slight increase in single cell
necrosis (apoptosis) seen in this study as well as in the
chronic 2-year findings where more regenerative repair
has been reported. There is a clear threshold for all of
these effects, supporting a threshold for the eventual
induction of liver cancer. The ACC study also provides
toxicokinetic data indicating the metabolism of 1,4-
dioxane shifts from linear, first-order metabolism to a
zero-order kinetics with increasing dose indicating
metabolic saturation. Once saturated, increased exposures
result in a disproportional increase in circulating levels of
1,4-dioxane. This supports a threshold response.

EPA considered the additional unpublished evidence
submitted by ACC but did not include this evidence in the
MOA analysis. The evidence in this unpublished report is not
sufficiently specific to provide support for a specific MOA.
While the study may identify thresholds for specific effects
evaluated in the study, a 90-day study that does not include
tumor endpoints is not able to demonstrate that the key events
in question are necessary precursors of liver tumor formation.

14, 58

MOA- General critiques of the MOA for liver
carcinogenicity proposed by Dourson et al.

Public Comments:

As described above, EPA has added a new Appendix I that
methodically applies the MOA framework outlined in the

Page 108 of 212

-------


•	Dourson et al., (2017) adjusted doses from sub-
chronic studies by dividing them by a factor of 3 in an
apparent attempt to compare them to values from
chronic studies. It is not appropriate to compare data
from different studies for the purpose of attempting to
define a quantitative relationship across studies by
adjusting doses for effects in sub-chronic studies.

•	Dourson et al., (2017)'s claim of a chronology of the
appearance of endpoints leading to liver tumors is
based on the appearance of liver pathology at different
doses, not over time. One cannot infer chronology
from dose-response data only.

•	The doses of 1,4-dioxane at which cytotoxicity and
cell proliferation were observed were greater than the
doses for tumor induction.

•	Dourson et al., (2017) relied on seven studies to argue
that their chronology of events resulted in tumor
formation. However, only three of the seven studies
provided tumor data.

Guidelines for Carcinogen Risk Assessment to evaluate
evidence for the MOA for liver tumors proposed by Dourson
etal., (2014, 2017).

EPA considered each of these critiques of the MOA proposed
by Dourson et al., in development of the MOA analysis in
Appendix I. EPA weighed the available evidence, including
each of the arguments articulated in these comments, and
concluded that there is insufficient evidence to support the
MOA proposed by Dourson et. al.. While some of the
available evidence is consistent with the proposed hypothesis,
alternate hypotheses cannot be ruled out. As discussed in
Appendix I, much of the evidence articulated in these
comments suggests that not all of the proposed key events are
necessary precursors to tumor formation. In addition, there
are several datagaps that prevent EPA from identifying causal
relationships between key events. For example, as noted by
commenters, there is no clear evidence demonstrating that
cytotoxicity is a necessary precursor to observed cell
proliferation.

58

MO A- Specific evidence for cytotoxicity and necrosis

as a key event leading to liver carcinogenesis

Public Comments:

•	Dourson et al., (2017) notes that tumors were found in
the low-dose group in the mouse study (Kano et al.,
2009), below the dose postulated to reflect saturation
kinetics. This is evidence that tumor formation and
non-tumor toxicity are decoupled. In the Kano et al.,
(2009) study, hyperplasia but not the postulated
intermediate step of necrosis/inflammation was seen
at 1,000 ppm.

•	Dourson et al., (2014) suggests that 1,4-dioxane

As described above, EPA has added a new Appendix I that
methodically applies the MOA framework outlined in the
Guidelines for Carcinogen Risk Assessment to evaluate
evidence for the MOA for liver tumors proposed by Dourson
etal., (2014, 2017).

EPA considered the available evidence for each key event in
the MOA for liver carcinogenicity proposed by Dourson et
al., including cytotoxicity.

Page 109 of 212

-------


causes liver tumors in rats and mice through a
pathway involving cytotoxicity (as indicated by
hypertrophy and necrosis) followed by regenerative
hyperplasia. This conclusion is not supported by the
data in female mice.

• Liver tumors identified from rodent liver bioassays
occurred in the absence of reported lesions related to
cytotoxicity (Kano etal., 2008; JBRC, 1998; NCI,
1978), suggesting that cytotoxicity may not be a key
event after 1,4-dioxane exposure leading to liver
carcinogenesis.

Based on the available evidence (including the evidence
outlined in these comments), EPA concluded that the
available evidence from Kano (2009) indicates increased
incidence of liver tumors at doses below those associated
with hepatocellular toxicity. This suggests that cytotoxicity
may not be a necessary key event in 1,4-dioxane exposure
leading to liver carcinogenesis. EPA's conclusion is
consistent with conclusions of these comments.

58

MOA- Evidence for hyperplasia as a key event in liver

carcinogenesis

Public Comment:

•	In the two-year drinking water study (Kociba et al.,
1971, 1974), hyperplasia/abnormal tumor foci were
not observed at a dose that did produce tumors.
"[CJritical intermediate steps (effects) in this causal
chain are missing even when subsequent steps are
observed, including at doses identified by Dourson et
al., (2017) as resulting in saturation kinetics."

•	In the NCI (1978) study, hyperplasia showed a dose-
response that significantly differed from that for
adenoma tumors. Relative to controls, the incidence of
hyperplasia dropped while adenoma incidence
increased. These data strongly suggest no significant
linkage between hyperplasia and adenomas.

As described above, EPA has added a new Appendix I that
methodically applies the MOA framework outlined in the
Guidelines for Carcinogen Risk Assessment to evaluate
evidence for the MOA for liver tumors proposed by Dourson
etal, (2014, 2017).

EPA considered the available evidence for each key event in
the MOA for liver carcinogenicity proposed by Dourson et
al., including hyperplasia.

Based on the available evidence (including the evidence
outlined in these comments) EPA concluded that the
available evidence suggests there may be a dose-response
relationship between 1,4-dioxane and bile duct epithelial
hyperplasia in Kociba (1974), but did not show a dose-
response relationship between 1,4-dioxane and hepatocellular
hyperplasia or demonstrate that hyperplasia precedes tumor
formation.

Page 110 of 212

-------
22, 58

MOA- Evidence from nasal tumors

Public Comment:

•	The key events for rat nasal tumor formation
(Kasai et al., 2009) show tissue injury with
regenerative repair (metaplasia) occurring in both
respiratory and olfactory mucosa.

•	The observed incidence of rare nasal tumor types
following 1,4-dioxane exposure is likely the result
of a genotoxic or mutagenic MOA and unlikely to
be attributed to a cytotoxic MOA (as evidenced by
other nasal carcinogens). Evidence for a genotoxic
or mutagenic MOA in one organ {i.e., the olfactory
system) should create a strong presumption that the
same MOA is operating in other organs {i.e., the
liver), absent compelling counterevidence.

EPA considered evidence for MO As in each tumor type
independently. There is insufficient evidence to support a
specific MOA or perform an in-depth MOA analysis for
tumor types reported in tissues other than liver. For tumor
types in the respiratory and olfactory mucosa, EPA
summarized possible MO As proposed by Kasai 2009 and
Kano 2009. As noted by SACC peer reviewers, the rare
nature of the nasal tumor types associated with 1,4-dioxane
exposure in rats indicates that the MOA for nasal tumors is
unlikely to be a generic cytotoxic/regenerative repair
response.

SACC

Toxicokinetics- Clarity of Discussion

SACC Recommendation: Improve the discussion of
toxicokinetics (Section 4.2.2) which the Committee found
confusing and vaguely worded.

EPA has substantially revised Section 4.2.2 on
toxicokinetics to use more precise language, incorporate
specific examples from experimental data, and provide more
quantitative information.

SACC

Toxicokinetics- Quantification

SACC Recommendation: Given there seem to be some
populations/situations involving chronic (repeated)
exposures, metabolism and elimination must be treated
quantitatively to determine temporal (spikes and troughs)
patterns in blood levels as part of the toxicokinetic
evaluation.

This would be possible if there was an adequate PBPK
model. EPA concluded an adequate PBPK model was not
available and could not be developed from available data.
EPA acknowledges that there is uncertainty around the
magnitude of variation in toxicokinetic factors across
individuals. EPA applied an uncertainty factor of 10 to
account for interindividual variability associated with
differences in toxicokinetic and toxicodynamics, though
there is uncertainty around whether the factor of 10 is
sufficient to protect potentially susceptible subpopulations.

SACC,
58

Acute Non-Cancer - Endpoint and Study Selection

EPA identified Mattie et al., 2012 as the best available
information to support an acute POD. Acute exposure

Page 111 of 212

-------
SACC Recommendation: Justify the use of Mattie etal.,
(2012) given its lack of critical detail (no histological
slides) and unclear quality control.

SACC Recommendation: Clarify how the relative
weight given the neurotoxicity studies of Mattie and
Goldberg was determined. An additional study by Kanada
(1994), which was apparently not considered in the
Evaluation, may need to be included in the neurotoxicity
systematic review and subsequent discussion.

Public Comment: The use of Mattie et al., (2012) for
derivation of an acute/short term inhalation point of
departure (POD) human equivalent concentration (HEC)
instead of the human study by Ernstgard el al., (2006)
which does not require interspecies extrapolation, is not
clearly justified. No assessment of data quality for this
study was presented to support its dismissal. It is
suggested that EPA use Ernstgard et al., (2006) as the
basis for its acute/short term inhalation modeling.

studies evaluating irritation and inflammation responses in
small numbers of human volunteers also contribute to the
overall weight of scientific evidence. The NOAEL for acute
irritation identified in short-term human studies indicates
that the acute POD (and corresponding benchmark MOE)
derived from the Mattie 2012 study is protective of acute
irritation in humans.

The Kanada 1994 study would have been filtered out in the
systematic review because it evaluates acute effects from a
single exposure and does not provide dose-response
information. While it identifies significant effects on
dopamine and serotonin following a relatively high dose and
provides qualitative evidence for neurotoxicity, it does not
provide information to support quantitative dose-response
analysis at lower levels of exposure. The Goldberg study
provided dose-response information on behavioral
outcomes, but the lowest dose evaluated in the study was
only half of the LD50. The Mattie study was selected as the
basis for the acute POD in part because it provides
quantitative dose-response information on effects at a lower
level of exposure.

The Ernstgard et al., 2006 paper was considered in the
overall weight of scientific evidence for acute hazard for
1,4-dioxane. EPA did not select it as the basis for the POD
because it is narrowly focused on evaluating acute irritation
and inflammatory response following just 2-3 hours of
exposure in a small sample of volunteers. Furthermore, it did
not identify any acute effects and therefore did not provide
quantitative information about dose-response beyond
identification of a NOAEL for a limited range of acute
outcomes. The Mattie study evaluates a panel of liver,	

Page 112 of 212

-------




kidney, and respiratory effects following a longer duration
of exposure in rats. The results of the Ernstgard study
(which identified a NOAEL of 72mg/m3) indicate that the
POD derived from the Mattie study (a POD of 284 mg/m3
with a benchmark MOE of 300 to account for uncertainty) is
protective of acute irritation or inflammatory effects in
humans

24

Developmental toxicity- discussing data gaps

Public Comment: EPA acknowledges that the database
for potential reproductive and developmental toxicity of
1,4-dioxane is deficient (p. 108), but later states it "did not
include women of reproductive age or pregnant women
who may work with 1,4-dioxane or children ages 16-21
because the acute effects on liver enzymes and CNS
effects are not expected to preferentially affect women or
developing children." EPA should not assume that a lack
of data is equivalent to a lack of risk.

EPA has rephrased the statement cited by the commenter to
be more clear that this decision is based on a lack of
information: "EPA does not have information to indicate that
the acute effects on liver enzymes and CNS effects would
preferentially affect women or developing children."

58

Developmental toxicity- filling data gaps

Public Comment: EPA should use its information
authorities to fill data gaps identified in the risk
evaluation, including dermal toxicity data and
reproductive/developmental/ neurodevelopmental toxicity
data.

As described in previous responses (see Section 1), EPA had
sufficient information to complete the 1,4-dioxane risk
evaluation using a weight of scientific evidence approach.
EPA selected the first 10 chemicals for risk evaluation based
in part on its assessment that these chemicals could be
assessed without the need for regulatory information
collection or development. When preparing this risk
evaluation, EPA obtained and considered reasonably
available information, defined as information that EPA
possesses, or can reasonably obtain and synthesize for use in
risk evaluations, considering the deadlines for completing the
evaluation. In some cases, when information available to
EPA was limited, the Agency relied on models; the use of
modeled data is in line with EPA's final Risk Evaluation Rule
and EPA's risk assessment guidelines.

Page 113 of 212

-------




EPA will continue to improve on its method and data
collection for the next round of chemicals to be assessed
under TSCA.

SACC

Nasal Effects- Evidence for systemic delivery

SACC Recommendation: Add a justification for
excluding all nasal effects in the extrapolation to dermal
routes from inhalation exposures. Delivery of 1,4-
Dioxane (or metabolite) in the blood stream may well
have contributed to the observed nasal effects. This is
documented by the widespread olfactory mucosal
distribution of lesions (which is typical of systemic
agents, not inhaled agent delivery). Also, nasal injury was
seen after oral exposures. Although aspiration of drinking
fluid might have occurred it can't be excluded as a cause.
More justification is needed.

EPA has reviewed the evidence of respiratory lesions and
concluded that these effects are consistent with systemic
delivery as opposed to exposure via portal of entry. EPA has
revised the dermal PODs extrapolated from inhalation
exposure for cancer and chronic noncancer endpoints to
incorporate nasal effects believed to be a result of systemic
delivery. EPA has also modified the discussion around these
PODs to reflect that conclusion. The revised chronic dermal
non-cancer POD is now based on nasal effects in inhalation
studies and is highly consistent (less than 2-fold difference)
with PODs derived from systemic effects following oral
exposures.

SACC

Nasal Effects- Characterization of Nasal Toxicology

SACC Recommendation: There were several issues with
the nasal toxicology that need to be addressed. For
example, the term "nasal" is too vague. More specific
language should be used to distinguish respiratory and
olfactory effects. The description of the nasal tumors
should include information on their distribution; this
could help define MOA. The document would be
strengthened by inclusion of a nasal toxicologist on the
writing team, and inclusion of a cogent discussion of the
nasal lesions relative to the current state of the art.

The original studies do not consistently provide the level of
specificity that would be required for detailed nasal
mapping. Where possible, EPA has revised language around
nasal lesions to be more specific. EPA has also revised
several PODs based on evidence for widespread distribution
of nasal lesions. As noted by SACC reviewers, the uniform
distribution of lesions (ie the lack of airflow gradient), in
combination with evidence of systemic circulation of 1,4-
dioxane following inhalation exposures, indicates that
lesions in the olfactory and respiratory epithelium can be
attributed to systemic delivery rather than portal of entry
effects.

Point (»r Departure ( 0111 incuts

SACC,
58, 59

PODs- BMD Modeling for Respiratory Metaplasia

EPA applied BMD modeling for all endpoints that were
amenable to modeling. Specific reasons that specific

Page 114 of 212

-------
SACC Recommendation: The benchmark modeling of
the respiratory metaplasia would benefit from additional
explanation and clarity. This includes explaining why a
high confidence benchmark dose (BMD) can be derived
from only two dose groups plus controls.

Public Comment: All endpoints within a study judged to
be relevant to the exposure should be considered when
modeling. This will help ensure that no endpoints with the
potential of having the most sensitive effect for risk
assessment applications, usually having the lowest
BMDL, are excluded from the analysis.

Public Comment: EPA needs to clarify its decisions to
use BMD modeling for some data sets, but not others
(e.g., modeling respiratory metaplasia but not atrophy of
the olfactory epithelium reported by Kasai etal., 2009).
BMD guidance suggests that neither data set meets the
minimum criteria for modeling since all of the non-
control exposures show a similar response level —
limiting any modelling of a dose-response relationship.
The response rate for both lesions at the lowest dose (68
and 80 percent) significantly exceeds the specified
benchmark response (BMR) of 10%. The large disparity
in the BMR and incidence rate at the lowest dose is a
strong indicator of lack of fit in the BMD modeling and
subsequent inappropriate BMD estimates.

The draft evaluation rejects a BMD approach for a mouse
data set with a very similar dose response. It is not clear
why this same logic was not applied to the evaluation of
nasal lesions from the rat studies.

Given the uncertainty introduced in applying the BMD
modeling of the respiratory metaplasia data, the Agency

endpoints were determined not to be amenable to BMD
modeling are stated in footnotes to each table. Additional
endpoint-specific explanations are provided in the BMD
modeling appendix (Appendix J). The LOAEC/NOAEC was
used for endpoints that were not amenable to benchmark dose
modeling.

As described in Appendix J, EPA modeled respiratory
metaplasia using only two doses because the highest dose
group was excluded to achieve better model fit. EPA does
not automatically consider data based on only 2 doses to be
concerning. Graphs and other results are evaluated on a
case-by-case basis for each endpoint, in particular for an
idea of how the results may be affected by choice of doses.
In this case it is seen that the fitted curve is superlinear in the
range of doses evaluated. Had there been additional testing
at lower doses, such testing might have revealed a
comparatively flat portion of the curve at low doses, which it
seems would have raised the BMDL. Thus, the BMDL
reported is probably below the true BMD. The BMDL is a
bound, not a point estimate. From a purely statistical
viewpoint, what it means to have confidence in the BMDL
is, we are reasonably comfortable assuming that the BMDL
is no larger than the population BMD. Additionally, in this
particular case, the BMDL of 4.7 ppm obtained from
modeling the respiratory metaplasia data with the high dose
dropped is virtually the same as what one would get using
the "fall back" LOAEC approach in which an additional
uncertainty factor of 10 is applied to the benchmark MOE to
account for LOAEC to NOAEC extrapolation (LOAEL
POD of 50 ppm/10 = 5 ppm).

Female mouse liver tumor data that was initially determined

Page 115 of 212

-------


should instead use the LOAEC to derive the HEC for
metaplasia. Since the LOAEC is the same as for atrophy,
the HEC for either lesion should be 17.6 mg/m3.
Although the dose response for the two endpoints is very
similar, there is a ten-fold difference in the HEC,
therefore (if modeling is going to be done), then both
endpoints should be modeled.

not to be amenable to modeling has since been incorporated
into the final draft using an alternate modeling approach.

58

PODs- LOAEC vs. BMD Approach

Public Comment: EPA states that the "LOAEC was used
with an uncertainty factor for LOAEC to NOAEC
extrapolation." However, where and how it did so is not
explained clearly in earlier sections of the document (e.g.,
Section 4.2.6.2.3, pp. 111-114), where text and tabular
calculations are provided for this outcome. It is unclear
why EPA included this text about the LOAEC to NOAEC
extrapolation within the discussion of Risk
Characterization Assumptions and Uncertainties (p. 150),
given that benchmark dose modeling (BMD) was used in
these calculations.

In the case of the chronic noncancer inhalation POD, there
were several endpoints that could not be modeled. In the
interest of transparency, EPA shows its quantitative dose-
response analysis for all relevant endpoints. The sensitive
endpoint that was ultimately selected as the basis for the
chronic POD used in quantitative risk calculations was
evaluated using BMD modeling. The discussion of
uncertainty related to the LOAEC and NOAEC in the Risk
Characterization Assumptions and Uncertainties (noted by
the commenter) was not relevant to the chronic endpoints
used in risk characterization and has been revised for clarity.

SACC,
58

PODs- Methodology for Calculating PODs for Portal
of Entry Effects

SACC Recommendation: Clarify why a flawed
Reference Concentration (RfC) methodology for portal of
entry effects is used in this Evaluation. The approach and
calculations for the inhalation POD appear to follow
standard EPA procedure and are calculated correctly
according to that procedure. However, it needs to be
recognized that the RfC procedure for portal of entry
effects itself is fundamentally flawed. It is based on faulty
assumptions and the RfC procedures provide dosimetry
estimates that are widely variant from actual experimental

In response to other SACC feedback, EPA reviewed
evidence in support of portal of entry vs. systemic effects.
Based on the location of lesions relative to airflow, and the
detection of 1,4-dioxane in blood following inhalation
exposure, EPA is now considering respiratory lesions
described in the literature to be primarily the result of
systemic delivery. EPA is therefore no longer calculating
any PODs for portal of entry effects in the revised RE. EPA
instead calculated PODs for chronic inhalation risk based on
assumptions about systemic effects.

To extrapolate dermal PODs from inhalation PODs, EPA
calculated human equivalent doses based on an inhalation

Page 116 of 212

-------


measurements. The EPA recognized this problem; a
subsequent review of the 1994 EPA RfC procedure
clearly described the inadequacy of the RfC protocol
[U.S.EPA (2009). Advances in inhalation dosimetry of
gases and vapors with portal of entry effects in the upper
respiratory tract (Vol. EPA/600/R-09/072). Washington,
DC],

Public Comment: EPA appears to use inappropriate
model inputs for the chronic non-cancer assessment for
dermal exposures extrapolated from chronic inhalation
studies (p. 117): The agency uses an inhalation rate of
1.25 mA3/hr for their inhalation to dermal conversion.
This does not match with the number in the EPA
Exposure Factors Handbook for average adult moderate
activity level (Table 6-28 suggests 2.1 mA3/hr). EPA
should explain the rationale for this deviation.

rate of 1.25m3/hr. as recommended in EPA's Engineering
Manual (Chemical Engineering Branch Manual For The
Preparation Of Engineering Assessments, 1991) That value
is based on a standard estimate that the typical worker
inhales 10m3 over the course of an 8 hour workday (see
REACH guidance on information requirements and
chemical safety assessment {ECHA, 2010, 6322478}). This
is the same breathing rate assumption that is used for
occupational exposure limits. This estimate of average
inhalation rate over the course of a workday reflects the fact
that work intensity varies substantially across the workday.
The hourly rates reported in the EPA Exposure Factors
Handbook do not easily translate to an 8 hr workday because
of these variations in work intensity over the course of a
shift. The daily average value of 1.25m3/hr is slightly above
NIOSH's estimated inhalation rate for light work
(1.18m3/hr) and below the NIOSH estimated inhalation rate
for moderate work (1.75m3/hr).

Note also that assuming a higher inhalation rate based on
moderate intensity work for the purposes of route-to-route
POD extrapolation would result in a higher POD that may
not be appropriate or adequately health protective for all
exposure scenarios.

SACC

PODs- Kasai et al., LOAEC

SACC Recommendation: Provide more justification for
the selection of 50 ppm as a LOAEC rather than as a
frank effect given the finding in the Kasai et al., (2009)
paper.

EPA does not make a distinction between a LOAEC and
frank effect in this risk evaluation. There is no functional
difference in how a LOAEC or a frank effect are used as the
basis for risk characterization.

SACC

PODs- 1,4-Dioxane Gas Category

SACC Recommendation: Given that this document is

Rather than apply default assumptions for a general gas
category, EPA has applied chemical-specific information
that informs the extent to which 1,4-dioxane toxicity is the

Page 117 of 212

-------


relying upon the RfC methodology, it is important that the
document explicitly state whether 1,4-Dioxane is viewed
as a category 1, 2 or 3 gas.

result of systemic delivery. EPA has substantially expanded
the discussion of the properties of 1,4-dioxane and
experimental observations (including its detection in blood,
systemic toxicity, and the uniform distribution of lesions in
olfactory and respiratory epithelium following inhalation
exposures) that provide insight into the role of systemic
delivery in its respiratory effects following inhalation
exposure (see Section 4.2.6.2.3). Based on these
observations, EPA treated 1,4-dioxane as a systemic acting
gas.

SACC,
24

Uncertainty Factors- Toxicokinetics

SACC Recommendation: Include the rationale for not
including a toxicokinetic uncertainty factor given the
toxicokinetic uncertainties associated with route to route
extrapolation.

Public Comment: EPA relied on oral-to-dermal
extrapolation (p. 90) for sub-chronic/chronic non-cancer
outcomes, with little acknowledgment of the substantial
uncertainties associated with route-to-route extrapolation.
The guidance cited for its extrapolation protocol explicitly
indicates the need for a thorough evaluation of
uncertainty, including "a qualitative evaluation of key
exposure variables and models, and their impact on the
outcome." Yet only a single statement of uncertainty —
"oral to dermal route-to-route extrapolation assumes that
the oral route of exposure is most relevant to dermal
exposures" is provided, which is far from sufficient.
Additional discussion of the uncertainty associated with
this extrapolation is needed. Prior research suggests the
inclusion of additional uncertainty factors for route-to-
route extrapolation may be appropriate. EPA should apply

A toxicokinetic uncertainty factor could be applied to
address uncertainty associated with route-to-route
extrapolation. A primary source of uncertainty related to
route-to-route extrapolation from inhalation exposures is the
relative efficiency of absorption through the lungs vs.
absorption through the skin. Absorption through lungs is
generally expected to be more efficient for solvents. The
dermal PODs derived under the assumption of 100%
absorption may therefore be artificially low but are unlikely
to be artificially high.

Another source of uncertainty in extrapolation from
inhalation exposures is that some of the key inhalation
studies used whole body exposures which could have
resulted in simultaneous exposure through other pathways
(eg oral exposure via grooming and vapor-through-skin).
These additional routes of exposure could result in an
underestimate of the amount of 1,4-dioxane require to
produce an effect in inhalation studies. If such alternate
pathways do contribute to total exposure in the inhalation
studies, the PODs derived from these studies (which assume
that all effects are due to inhalation exposure alone) could be
artificially low.

Page 118 of 212

-------


an additional uncertainty factor of 10 to account for these
uncertainties.

The primary source of uncertainty related to oral-to-dermal
extrapolation is the difference in kinetics related to first pass
metabolism. It is unknown whether 1,4-dioxane or its
metabolites are the primary source of toxicity, or whether
metabolism in the skin and other tissues produce unique
metabolites with higher or lower toxicity. While these are
important source of uncertainty for most tissues,

In this risk evaluation, EPA performed oral-to-dermal
extrapolation for liver endpoints (an HED based on
hepatocellular toxicity and a CSF based on liver tumors in
female mice). Given first pass metabolism, it is unlikely that
dermal exposure would result in greater exposure to the liver
than oral exposures would.

Because the sources of uncertainty related to route-to-route
extrapolation in this risk evaluation all contribute to a
potential underestimate of the POD, extrapolation from
inhalation or oral to dermal exposure is expected to be a
relatively conservative approach. EPA concluded that it is
not necessary to apply an additional uncertainty factor.

24, 58

Uncertainty Factors- Database Uncertainty
Public Comments:

• EPA fails to include necessary uncertainty factors in
its calculations of benchmark margins of exposure
(BMOE) for risks to workers of non-cancer effects
from inhalation and dermal exposure. The BMOE that
EPA derives is 30, resulting from the multiplication of
two uncertainty factors—3 for interspecies variation
(UFA) and 10 for intraspecies variation (UFH). EPA
should include an additional uncertainty factor for "the

There is no universal list of hazard data required when
evaluating chemical risks under TSCA. Furthermore, for
1,4-dioxane, EPA has sufficient, reasonably available hazard
information to conduct a risk evaluation and support the use
of the chosen hazard endpoints. Therefore, EPA did not use
a database uncertainty factor for hazard in the 1,4-dioxane
risk evaluation.

Page 119 of 212

-------


uncertainty associated with extrapolation from animal
data when the database is incomplete."

•	With regard to the database uncertainty factor, the
EPA Risk Assessment Forum notes in its 2002 report,
A Review of the Reference Dose and Reference
Concentration Processes: The database UF is intended
to account for the potential for deriving an under-
protective RfD/RfC as a result of an incomplete
characterization of the chemical's toxicity.

•	The 1,4-dioxane database is clearly incomplete. There
is no dermal toxicity data at all and only a single short-
term developmental toxicity study; hence, the Agency
lacks any sub-chronic or chronic reproductive,
developmental or neurotoxicity data. Thus, it is
imperative that EPA apply an additional uncertainty
factor of 10 to account for these data gaps.



59

Cancer- Selection of relevant endpoints

Public Comment: Evidence suggests that peritoneal
mesothelioma and mammary gland adenomas presented in
drinking water studies were spontaneous tumors and likely
not appropriate for inclusion in the risk evaluation.

EPA included these tumor types in the MS-combo cancer
models because dose-response data indicate an association
between exposure to 1,4-dioxane and increased incidence of
these tumor types. This is consistent with EPA cancer
guidelines, which state that, "The default option is that
positive effects in animal cancer studies indicate that the
agent under study can have carcinogenic potential in
humans".

14, 24,
53

Cancer- Kano (2009) data on liver tumors in female
mice

Public Comment: EPA should reconsider evidence for
tumors in mice and re-evaluate the threshold for cancer in
workers to align with prior agency practice. Specifically,
the female hepatocellular carcinoma data from Kano et

In response to these comments, EPA revised the dermal
cancer slope factor (extrapolated from effects observed in
oral exposure studies) to incorporate the sensitive liver cancer
endpoints observed in female mice in Kano 2009. EPA
revised associated risk calculations and tables to reflect this
change. While the data were initially excluded because they
are not well suited to modeling, EPA was able to collect

Page 120 of 212

-------


al., (2009), which was chosen as the basis for the USEPA
IRIS oral cancer slope factor, was disregarded. The EPA
rationale is not scientifically supportable, and it is
recommended that the IRIS oral cancer slope factor be
used as the oral cancer slope factor in the risk evaluation;
the IRIS CSF is more stringent than the chosen value.
Exclusion of this data results in an oral CSF of 0.021
(mg/kg/day) that is approximately 5-fold less protective
than the CSF identified the 2013 IRIS assessment. Table
4-12 (pg. 126) must include hepatocellular tumors
observed in female mice in the Kano et al., (2009) study.

additional individual animal data from the study authors to
support a time-to-tumor analysis that provides a stronger
basis for modeling. The new approach to modeling the
female liver tumor data resulted in a revised oral (and
dermal) CSF of 0.1 (mg/kg/day)-1.

24, 56,
48

Cancer- Apply Cancer Guidelines to selection of linear

vs. threshold approach

Public Comments in Support of the Linear Approach

•	According to agency cancer guidelines, the agency
shall use the default linear approach in the absence of
an alternative known MO A. Only when "alternative
approaches have significant biological support"
should an "assessment...present results using
alternative approaches".

•	There is a scientific consensus that there is insufficient
evidence to support a threshold approach and
therefore, the EPA should follow agency guidelines
and not entertain the development of a threshold
model.

•	Human variability with respect to the individual
thresholds for a nongenotoxic cancer mechanism can
still result in linear dose-response relationships in the
population.

Public Comments in Support of a Threshold Approach

•	EPA's cancer guidelines state: If critical analysis of

EPA's cancer guidelines state that "When the weight of
evidence evaluation of all available data are insufficient to
establish the mode of action for a tumor site and when
scientifically plausible based on the available data, linear
extrapolation is used as a default approach, because linear
extrapolation generally is considered to be a health-
protective approach. Nonlinear approaches generally should
not be used in cases where the mode of action has not been
ascertained. Where alternative approaches with significant
biological support are available for the same tumor response
and no scientific consensus favors a single approach, an
assessment may present results based on more than one
approach" (p.3-21 of the guidelines).

For 1,4-dioxane, EPA concluded that there is insufficient
evidence to determine whether 1,4-dioxane or its metabolites
act through a mutagenic or otherwise genotoxic mode of
action. EPA also reviewed evidence for several plausible
MO As and concluded that there is insufficient evidence to
determine the mechanism of action for carcinogenicity of
1,4-dioxane for any of the tumor locations. While some

Page 121 of 212

-------


agent-specific information is consistent with one or
more biologically based models as well as with the
default option, the alternative models and the default
option are both carried through the assessment and
characterized for the risk manager.

evidence for the proposed MOA of metabolic saturation
followed by cytotoxicity and regenerative proliferation is
available for liver tumors, the available evidence is also
consistent with alternate plausible MO As (as outlined in
Appendix I). Applying a threshold approach to evaluate
cancer risk for 1,4-dioxane would not be adequately
supported by available mechanistic evidence. Consistent
with EPA guidance, EPA performed BMD analysis on
tumor data and applied the best fit models for the data.

48, 56

Cancer- Use MOA evidence to inform selection of a
linear vs. threshold approach

Public Comment in Support of a threshold approach
based on MOA evidence:

•	EPA failed to fully consider all of the available
evidence for key events supporting a threshold for
carcinogenic response in exposed animals

•	The conclusion that evidence for a threshold mode of
action was not sufficiently robust is incorrect.

•	The weight of evidence clearly supports that the mode
of action for rodent tumors associated with high doses
of 1,4-dioxane does not include the potential for
mutagenicity, and the science clearly supports a
threshold for both noncancer and cancer effects.
Therefore, linear extrapolation was inappropriate.

Public Comment in Support of a linear approach

based on MOA evidence:

•	The practice of assigning "nonlinear" MO As does not
account for mechanistic factors that can create
linearity at a low dose, such as when an exposure
contributes to an existing disease process.

As described above, EPA has added a new Appendix I that
methodically applies the MOA framework outlined in the
Guidelines for Carcinogen Risk Assessment to consider all of
the available evidence for the MOA for liver tumors
proposed by Dourson et al., (2014, 2017).

EPA concluded that there is insufficient evidence to
determine whether 1,4-dioxane or its metabolites act through
a mutagenic or otherwise genotoxic mode of action. EPA also
reviewed evidence for several plausible MO As and
concluded that there is insufficient evidence to determine the
mechanism of action for carcinogenicity of 1,4-dioxane for
any of the tumor locations. While some evidence for the
proposed MOA of metabolic saturation followed by
cytotoxicity and regenerative proliferation is available for
liver tumors, the available evidence is also consistent with
alternate plausible MO As (as outlined in Appendix I).
Applying a threshold approach to evaluate cancer risk for
1,4-dioxane would not be adequately supported by available
mechanistic evidence. Consistent with EPA guidance, EPA
performed BMD analysis on tumor data and applied the best
fit models for the data.

Page 122 of 212

-------
48, 59,

Cancer- Specific Comments in Support of a Non-

Linear Threshold Approach

Public Comments:

•	EPA should consider the Health Canada approach for
the assessment of 1,4-dioxane in drinking water. This
approach, which has been informed by Meek et al.,
(2014) ultimately considers the cancer and non-cancer
effects of 1,4-dioxane together using a threshold
approach.

•	Evidence for a threshold effect caused by
accumulation of the parent compound is strongest for
liver tumors.

•	Nasal squamous cell carcinoma, peritoneal
mesothelioma, subcutis fibroma, and hepatocellular
adenoma or carcinoma tumor incidence data from
Kano et al., (2009) and Kociba et al., (1974) were
reported to be "clearly associated with a threshold"
(also supported by Torkelson et al., 1974), and should
not have been applied to a linear model.

As stated above, EPA concluded that there is insufficient
evidence to determine whether 1,4-dioxane or its metabolites
act through a mutagenic or otherwise genotoxic mode of
action. Consistent with EPA guidance, EPA performed BMD
analysis on tumor data and applied the best fit models for the
data.

•	Rather than follow the Health Canada approach, EPA
applied an approach consistent with its own guidelines for
carcinogen risk assessment.

•	While the mechanistic evidence is strongest for liver
tumors, EPA concluded there is not sufficient evidence to
support a specific MOA for liver carcinogenicity.

•	While there is some indication of an apparent threshold
for tumor incidence in animal studies, EPA considered all
of the available evidence, including dose-response
information across all studies and tumor locations and
mechanistic information to inform model selection. EPA
performed a sensitivity analysis to quantify the impact of
applying a linear vs. threshold model for liver tumors on
overall cancer risk estimates.

58, 24,
56

Cancer- Specific Comments in Support of a Linear

Approach

Public Comments:

•	The most compelling argument for retaining the U. S.
EPA default assumption of linearity for 1,4-dioxane is
the presence of multiple tumor types in rodent models,
all of which are relevant to humans.

•	Although the linear non-threshold model for
carcinogenicity is the correct choice, language in the
Risk Evaluation sows doubt on its by stating that there
was a high degree of uncertainty in all of the MOA

•	EPA agrees that assuming linearity is appropriate given
the lack of information about MOA for several of the
tumor sites. EPA also used the MS-combo model to
perform a sensitivity analysis to determine the impact of
applying a linear vs. threshold approach for liver tumors
on overall cancer risk.

•	In the revised risk evaluation, EPA performed an in depth
MOA analysis (Appendix I) to demonstrate that all
evidence had been considered in reaching the decision to

Page 123 of 212

-------


hypotheses considered in the evaluation (e.g.,
mutagenic mode of action or threshold response to
cytotoxicity and regenerative hyperplasia for liver
tumors).

•	With regard to a potential threshold based on enzyme
saturation, it is well documented that enzymatic
metabolic activity varies across the population.
Therefore, it is inappropriate to assume that a possible
threshold found in limited in vivo studies in laboratory
animals or in vitro studies apply across the entire
distribution of the human population. Even if there
were a threshold seen in such studies based on
metabolic saturation, EPA would need to consider
variation in the human population and protect the
most sensitive individuals, who may experience this
purported "threshold" at lower doses.

•	In animal tests, a specific chemical may cause cancer
through a nonlinear dose-response process, but for the
human population, the dose-response relationship for
the same chemical is likely a low-dose linear one,
given the high prevalence of pre-existing disease and
background processes that can interact with a
chemical exposure, and given the multitude of
chemical exposures and high variability in human
susceptibility.

apply a linear model and to ensure transparency in the
process. EPA concluded that there is uncertainty in all of
the MOA hypotheses because there is in fact insufficient
information to decisively identify a particular MOA of
carcinogenesis for any tumor location.

•	EPA concluded that there was insufficient evidence to
conclude that metabolic saturation is a key event in liver
carcinogenesis for 1,4-dioxane. The final cancer risk
calculations for 1,4-dioxane do not rely on a threshold
approach and therefore do not require adjustments for
sensitive individuals for whom the threshold may be
lower.

•	EPA has considered the potential for pre-existing disease
and other factors that aren't reflected in animal studies
and may make some people more susceptible to both the
cancer and non-cancer effects of 1,4-dioxane (see
discussion on potentially exposed and susceptible
subpopulations in Section 4.4). The final cancer risk
calculations do not a rely on a threshold approach.

23, 59

Cancer- BMD Modeling

Public Comment: EPA needs to clarify why MS-Combo
was applied twice to tumor data to evaluate uncertainties
related to model choice and mechanisms.

• Rationale for the decision to choose default linear low-
dose extrapolation to develop a unit risk estimate and

EPA ran MS-Combo with and without liver tumors included.
This was performed as a sensitivity analysis to determine the
impact of including liver tumors on overall cancer risk. For
inhalation, EPA found that excluding liver tumors had little
impact on overall model results. This means that applying a
threshold model based on alternate MOA conclusions for
liver tumors would not substantially alter overall inhalation

Page 124 of 212

-------


cancer slope factor based on benchmark dose
modeling for multiple tumor sites is inadequately
described.

•	Use of MS-Combo requires that data sets exhibit a
statistically or biologically significant dose-related
trend. Neither is true for the following tumor types-
Zymbal gland adenoma, renal cell carcinoma, and
subcutis fibroma. While the increase in peritoneum
mesothelioma and mammary gland adenoma is
statistically significant, the biological significance of
these tumors to humans is questionable.

•	Given the failure of many of the tumor types to meet
the criteria for benchmark dose modeling, significant
uncertainty in interpreting multi-tumor analysis, and
since almost all of the target organs have an increased
tumor incidence only at the highest dose, the use of the
MS-Combo model for assessing the cancer risk of 1,4-
dioxane is inappropriate.

•	The importance of access to individual animal data
versus aggregated tumor dose-response data should be
investigated. Without access to individual animal data,
it is not known whether the same or different rats have
developed tumors in the various target organs. If the
same subset of rats has developed multiple tumors, one
might draw a different conclusion than if they were
observed in different rats and MS-Combo should not
be applied.

•	No justification is provided for choosing an alternative
threshold (a = 0.05) for the cancer model.

cancer risk conclusions. This has been clarified in the dose-
response section.

EPA assumed that different tumor types are independent and
used the MS-Combo model to characterize total cancer risk
in studies where 1,4-dioxane increased incidence of multiple
tumor types. This approach is consistent with NRC
conclusions that an approach based on counts of animals with
one or more endpoints would tend to underestimate risk when
tumors occur independently across sites (NRC. 1994).

Individual animal data showing that multiple tumors occur in
the same animal would not provide clear, sufficient
information to determine whether or not the tumors arose
independently.

59

Cancer- MS-Combo Model

The MS-Combo model was peer reviewed in 2011. As noted
on EPA's website. "The peer reviewers were specifically

Page 125 of 212

-------


Public Comment: Peer review of the application of the
MS-Combo model is needed before this model is used in
chemical-specific risk assessments. The EPA draft
indicates that the MS-Combo module has been peer-
reviewed. In fact, however, the first sentence of the 2011
report emphasizes that documentation provided to users is
clear enough to be adequate to allow users to run the
program and obtain program output, but is not adequate to
inform users concerning details about the context in which
applying the model is appropriate or intended.

asked to evaluate MS-Combo with respect to clarity of the
documentation and model output, and the adequacy of testing
methods and test results. The reviewers generally agreed that
the model produced statistically valid results but made
several recommendations regarding enhancements that could
facilitate or expand its practical application, and how the
documentation and outputs could be improved with respect to
clarity. EPA revised the MS-Combo software and
documentation in response to these comments, incorporating
most of the suggested revisions." While the model itself may
not provide instructions about the context in which applying
the model is appropriate or intended, EPA modelers can
evaluate this on a case-by-case basis. EPA also relies on the
SACC to provide peer review of its methods. SACC peer
reviewers did not raise concerns about the validity of EPA's
application of the MS-combo model for 1,4-dioxane.

56

Uncertainty- Interindividual Variability

Public Comment: The NAS has recommended that
human variability in response to carcinogens be accounted
for. A factor of 25- to 50- may account for the variability
between the median individual and those with more
extreme responses, and 25 was recommended as a
reasonable default value.

EPA evaluated cancer risk from 1,4-dioxane using an
approach consistent with the EPA Guidelines for Carcinogen
Risk Assessment.

24

Uncertainty- BMD Modeling

Public Comment: A discussion of assumptions and
uncertainties relevant to BMD should be provided

In response to this comment, EPA has inserted additional
discussion of strengths and limitations of BMD modeling
approaches for each endpoint into a new section describing
"overall confidence" in PODs (Section 4.2.7) as well as in
the "assumptions and uncertainties" section of the Risk
Characterization (Section 5.3). There is also detailed
discussion of BMD modeling assumptions for each endpoint

Page 126 of 212

-------




in Appendix J.

56

Uncertainties- Susceptible Populations/Human
Variability in Cancer Risk

Public Comment: California EPA reviewed the evidence
on differential susceptibility to carcinogens based on age
and life stage and derived age adjustment values for
carcinogens which include the prenatal period, proposing
a default Age-Sensitivity Factor of 10 for the third
trimester until age 2 years, and a factor of 3 for ages 2
through 15 years to account for potential increased
sensitivity to carcinogens during childhood. At a
minimum, EPA should incorporate factors to account for
human variability in response to carcinogens, as well as
Cal EPA's age adjustment values to address these known
susceptibilities. Uncertainty factors should be applied to
account for susceptibility of certain subpopulations based
on pre-existing health conditions.

Current EPA guidance recommends application of age-
dependent adjustment factors (ADAFs) only for carcinogens
shown to act through a mutagenic mode of action. For 1,4-
dioxane, EPA concluded that there is insufficient information
to determine whether a mutagenic mode of action contributes
to carcinogenicity.

SACC

General- Clarification Regarding NIOSH Criteria

SACC Recommendation: Clarify the text on page 153
regarding the entry for NIOSH (2017). As written this
entry could be interpreted to suggest NIOSH developed
its criteria document and the Recommended Exposure
Limit (REL) for 1,4-Dioxane using a linear 10-4
theoretical excess cancer risk level—the REL was derived
in 1977, which is 5 years before publication of the Howe
and Crump (1982) GLOBAL82 method report to OSHA.

EPA has modified this text to make clear that it is referring
to general NIOSH guidance: "For 1,4-dioxane, EPA used 1
x 10"4 as the benchmark for the purposes of this
unreasonable risk determination for individuals in industrial
and commercial work environments subject to Occupational
Safety and Health Act (OSHA) requirements. This cancer
benchmark is consistent with guidance outlined in the 2017
NIOSH chemical carcinogen policy."

EPA has also modified the accompanying footnote to add
the following "Note that the NIOSH REL for 1,4-dioxane
was established prior to this guidance."

SACC

General- Defining Qualifying terms

SACC Recommendation: Explicitly define qualifying

Where possible, EPA has replaced vague words with more
precise language and/or supported them with specific
quantitative information. For example, EPA substantially

Page 127 of 212

-------
terms used throughout the text (e.g., "acceptable," "high,
"extensive," "appreciable accumulation," and "rapid").

revised the section on toxicokinetics to include more specific
experimental observations on the "extent" of absorption and
metabolism.

The term "acceptable" is used in reference to study quality.
As described in the human health hazards "approach and
methodology" section of the risk evaluation, EPA uses a
quantitative scoring system to rate studies as high, medium,
low, or unacceptable. "Acceptable" study quality means that
the study met minimum inclusion criteria (ie it was not
"unacceptable") when EPA applied the systematic review
protocol.	

Page 128 of 212

-------
6. Risk Characterization

Charge Question 6.1: Please comment on the objectivity of the underlying data used to support the risk characterization and the
sensitivity of the agency's conclusions to analytic assumptions made.

Charge Question 6.2: Please comment on the characterization of uncertainties and assumptions including whether EPA has
presented a clear explanation of underlying assumptions, accurate contextualization of uncertainties and, as appropriate, the
probabilities associated with both optimistic and pessimistic projections, including best-case and worst-case scenarios. Please
provide information on additional uncertainties and assumptions that EPA has not adequately presented.

Charge Question 6.3: Please comment on whether the information presented supports the findings outlined in the draft risk
characterization section. If not, please suggest alternative approaches or information that could be used to develop a risk finding in
the context of the requirements of the EPA's Final Rule, 1'ioeedures for Chemical Risk Evaluation Under the Amended Toxic
Substances Control Act (i 16).

Summary of Comments lor Specific
Issues Related lo Charge Question 6

KPA/OPPT Response

SACC,
46, 55,
58

Data Gaps and Scope

SACC Recommendation: Clarify portions of text as

indicated

Specific examples in need of clarification identified

by committee members:

• limited data sets from the EU risk assessment

• monitoring data lacked descriptions of worker
tasks, exposure sources, and possible engineering
controls, assumed as personal breathing zone (PBZ)
measurements, sampling rate missing for some
2016 data

• the Agency recognizing some data sources may be
biased

• uncertainty on the underlying exposure distribution

• take home exposure from workers to family

Where possible EPA has clarified language related to these
issues.

Several of the examples of topics requiring clarification
identify data gaps. EPA is charged with performing risk
evaluations based on the best available science. EPA believes
it has sufficient information to make a reasoned analysis
regarding 1,4-dioxane in the limited time available for
completing the risk evaluation. EPA recognizes that limited
data for some exposure scenarios, hazard endpoints and
mechanisms introduces additional uncertainty around final risk
conclusions. These uncertainties are acknowledged in the
uncertainties sections (Section 5.3).

• The limitations of the EU Risk Assessment data sets are
discussed in the Key Uncertainties section of the risk
evaluation.

Page 129 of 212

-------
members was not considered
• There are what appear to be inconsistencies or
flawed assumptions in the discussion of the
estimates of number of workers exposed (page
146):

o "Furthermore, market penetration data was
unavailable, therefore, EPA was unable to
estimate the number of establishments within
each NAICS code that used 1,4-Dioxane
instead of other chemicals. This would result in
a systematic overestimation of the count of
exposed workers." The assumption of
overestimation in the count of exposed workers
is not self-evident. It could be more
establishments actually use the chemical and
thus more workers are actually exposed,
o "Second, EPA's judgments about which
industries" (represented by North American
Industrial Classification System (NAICS)
codes) and occupations (represented by
Standard Occupational Classification (SOC)
codes) "are associated with the uses assessed in
this report are based on EPA's understanding of
how 1,4-Dioxane is used in each industry.
Designations of some industries/occupations
with few exposures might erroneously be
included, or some industries/occupations with
exposures might erroneously be excluded. This
would result in inaccuracy but would be
unlikely to systematically either overestimate or
underestimate the count of exposed workers."
There does not appear to be enough evidence to
make the case that number of workers would

EPA assumes this comment is in relation to the BASF
manufacturing data based on context. The limitations and
uncertainties of this monitoring data and the monitoring
data for all other occupational exposure scenarios are
discussed in the Key Uncertainties sections of the risk
evaluation.

EPA acknowledges the potential for increased bias in some
of the monitoring data, especially that which was
promulgated from employee complaints. Discussion of bias
is included in the Key Uncertainties sections of the risk
evaluation.

EPA recognizes the uncertainty attached to the data it uses
in the Key Uncertainties sections of the risk evaluation.

Take home exposure was not addressed as it was defined
out-of-scope for the risk evaluation.

In the absence of market penetration data for a given
condition of use, EPA assumed 1,4-dioxane may be used at
up to all sites and by up to all workers calculated in the
method described in Appendix F.5 of the RE. In cases
where this approach is used, the number of workers
estimated is considered a bounding estimate with the actual
number of workers exposed being less than this estimate.
EPA has reworded the final phrase to better reflect the
confidence of the cited statement. Revision reads: "... This
would result in inaccuracy but is not expected to
systematically either overestimate or underestimate the
count of exposed workers." This revision, though minor,
removes the word "unlikely," which could be interpreted as
a qualitative representation of statistical odds, which EPA
does not know.

o EPA acknowledges that in some cases, ONU exposures
may be higher than worker exposures because of
unusual facility layout but expects this to be the	

Page 130 of 212

-------
not be systematically influenced,
o "Exposure data for ONUs were not available
for most scenarios. EPA assumes that these
exposures are expected to be lower than worker
exposures, since ONUs do not typically directly
handle the 1,4-Dioxane nor are in the
immediate proximity of 1,4-Dioxane. Only
inhalation exposures to vapors are expected,
which will likely be less than worker
exposures." However, given that inhalation is
the primary route of exposure, such an
assumption requires knowing something about
the layout of the facilities and how close ONUs
may actually be to where 1,4-Dioxane is being
used by workers. Given that this is a vapor, it is
likely not to be contained to direct use areas.

•	The Evaluation shows bias in its discussion of
potential bias: "Some data sources may be
inherently biased. For example, NIOSH HHEs for
the open system functional fluids and film cement
uses were conducted to address concerns regarding
adverse human health effects reported following
exposures during use. Both HHEs were requested
by relevant workers' unions (United Paperworkers
International Union and Film Technicians Union,
respectively)." Industry monitoring data are
possibly biased towards lower values, via, e.g.,
repeated measurements or representativeness, as
also cited in the limitations.

•	There are serious limitations regarding the exposure
data, as cited in the section on limitations: "The
95th and 50th percentile exposure concentrations
were calculated using reasonably available data.

exception. In most cases ONU exposure will be less
than a worker because ONU work tasks are more distant
to the inhalation source compared to those directly
working with 1,4-dioxane.

EPA recognizes the potential for bias in industry data. The
cited NIOSH HHE was used as an example to show bias but
is not presented as the only incidence of bias in the risk
evaluation. EPA has reviewed and added a more balanced
discussion on bias to underscore the potential for bias in all
areas, not just with HHEs and reported issues.
EPA understands the limitations of the data and the
resulting limitations of the risk evaluation. The limitations
are discussed in many places within the risk evaluation.
During systematic review, EPA determined that there were
nine (9) acceptable studies that characterized the aquatic
toxicity of 1,4-dioxane. These happen to be the same studies
that were used in previous risk assessments for 1,4-dioxane.
As a result of EPA's systematic review process, these
studies were rated "acceptable," "high," or "medium"
quality. EPA only evaluated studies that were "acceptable"
and rated as "high," "medium," or "low" quality. EPA
evaluated two acceptable studies that reported chronic
toxicity to fish. Other studies that addressed chronic
ecotoxicity were not acceptable for this assessment. In
addition, EPA applied assessment factors to derive the
environmental concern levels in the environment. These
assessment factors provided a lower bound effect level that
would likely encompass more sensitive species that are not
specifically represented by the reasonably available
experimental data. Assessment factors account for
differences in inter- and intra-species variability, as well as
laboratory-to-field variability. Additional data for chronic
ecotoxicity would only reiterate the conclusion the hazard

Page 131 of 212

-------
The 95th percentile exposure concentration is
intended to represent a high-end exposure level,
while the 50th percentile exposure concentration
represents typical exposure level. The underlying
distribution of the data, and the representativeness
of the available data, are not known." Recognizing
the limitations, this nonetheless raises serious
questions about the risk characterizations for the
inhalation exposures overall.

•	limited chronic toxicity studies available for
assessing the long-term effects to aquatic species

•	only one developmental study available

•	high degree of uncertainty in the MOA evaluations
in general

•	skin sensitization and respiratory sensitization were
not considered

•	Uncertainties include the provision of unmeasured
exposure concentrations in the Bringman and Kuhn
(1977) study. It is also uncertain how "intoxication"
is defined for that study. While the fathead minnow
study provided data regarding hatch and
development which were used in the chronic
threshold, no fish reproduction study could be
found. The lack of reproduction assessment
suggests significant uncertainty and may require
implementation of an additional safety factor.

Public Comment

•	EPA's risk determinations for 1,4-dioxane are
weakened by numerous data gaps that should have
been identified and addressed before initiating the
evaluation.

•	EPA's occupational exposure assessment is not

of 1,4-dioxane is low. Therefore, EPA is confident that 1,4-
dioxane does not pose any hazard resulting from chronic
exposures.

EPA has acknowledged the limited developmental toxicity
data as a source of uncertainty

EPA has substantially expanded the discussion of the MOA
in a new appendix and acknowledges the lack of clear MOA
as a source of uncertainty. EPA also includes a sensitivity
analysis showing that exclusion of liver tumors that may
occur through an alternate MOA does not substantially
change the cancer risk estimates for inhalation exposures
EPA considered reasonably available hazard data and
acknowledged the uncertainties associated with data gaps.
EPA did not find where "intoxication" was reported for any
studies that were evaluated by Bringman and Kuhn (1977).
For Daphnia magna, behavior, equilibrium and
immobilization were reported as effect to acute exposure to
1,4-dioxane. For algae, population growth was used to
summarize the effects.

Page 132 of 212

-------


supported by sufficient inhalation or dermal
exposure data.

• EPA's reliance on summaries of foreign exposures
does not provide substantial evidence for EPA's
determinations of no reasonable risk or satisfy
EPA's TSCA obligation to consider all reasonably
available information.

o Courts have previously held that foreign
studies of occupational exposure and risk
"do not constitute substantial evidence for
OSHA's finding of a significant risk," due
to the differences in working conditions
between the countries where the studies
were conducted and the United States.



SACC,
24, 46,
50, 54,
58

Data Gaps- Use authority to obtain additional data

SACC Recommendation: The Agency should use its
authority to obtain additional data to fill data gaps
and/or perform a quantitative sensitivity analysis. This
could further direct the Agency to areas where
additional important data are to be obtained, clarified,
and/or to apply a quantitative uncertainty analysis.

•	due to the lack of rigorous uncertainty and
sensitivity analysis, it is difficult to know whether
filling certain data gaps would make a material
difference to the Agency's conclusions.

Public Comment:

•	Qualitative and screening-level environmental
assessments were conducted where data are lacking
(e.g., sediment, land-applied biosolids). TSCA
requires EPA to conduct risk evaluations of

As described in previous responses (see Section 1), EPA had
sufficient information to complete the 1,4-dioxane risk
evaluation using a weight of scientific evidence approach. EPA
selected the first 10 chemicals for risk evaluation based in part
on its assessment that these chemicals could be assessed
without the need for regulatory information collection or
development. When preparing this risk evaluation, EPA
obtained and considered reasonably available information,
defined as information that EPA possesses, or can reasonably
obtain and synthesize for use in risk evaluations, considering
the deadlines for completing the evaluation. 40 CFR 702.33. In
some cases, when information available to EPA was limited,
the Agency relied on models; the use of modeled data is in line
with EPA's final Risk Evaluation Rule and EPA's risk
assessment guidelines. EPA will continue to improve on its
method and data collection for the next round of chemicals to
be assessed under TSCA.

Page 133 of 212

-------


chemicals that are based on the "best available
science" (TSCA section 26(h)) and all "reasonably
available information" on hazards and exposures
(TSCA section 26(k)). EPA did not use its
authority to reasonably generate, obtain, or
synthesize data for use in risk evaluations."



SACC,
22, 58

Assumptions and Uncertainties- General/Overall
Risk

SACC Recommendation: Make it more transparent
where uncertainties are quantified and provide
justification where they are not

SACC Recommendation: Provide more explanation
where requested.

•	Committee members wanted more uncertainty and
sensitivity analysis to ascertain whether the
underlying data are sufficient for the risk
characterization; One Committee member provided
references on quantifying uncertainties to inform
decision making (NAS 2014, Simon et al., 2016)

•	What information does EPA consider to be
"reasonably available?" That phrase is used
throughout the document without adequate
explanation.

•	Some of the assumptions and the resultant
uncertainty factors require more detailed
explanation. Examples include dermal absorption
fractions and interspecies uncertainty factors
(UFA). Considering statements about the lack of
data in some of the human studies, despite some

•	To the extent possible, EPA has tried to quantify
uncertainty where possible and to identify sources of
uncertainty where quantification is not possible.

•	As described above, reasonably available information is
defined as information that EPA possesses, or can
reasonably obtain and synthesize for use in risk
evaluations, considering the deadlines for completing the
evaluation. 40 CFR 702.33.

•	EPA has inserted additional discussion of the uncertainty
factors that were applied in the risk characterization

•	EPA has modified the discussion of uncertainties related to
dermal risk characterization to remove misleading language
and reflect changes in our approach.

The discussion of uncertainty related to the LOAEC and
NOAEC in Section 5.2.4 on Risk Characterization
Assumptions and Uncertainties (noted by the SACC) was
not relevant to the endpoints used in risk characterization
and has been revised for clarity.

•	EPA has rephrased the statement about risks to pregnant
women and children (ages 16-21) to clarify that there is a
lack of data on specific effects in these groups: "Workers
were identified as relevant potentially exposed or
susceptible subpopulations, but EPA did not specifically
identify women of reproductive age or pregnant women

Page 134 of 212

-------
degree of correspondence between effects observed
in humans and experimental animals, the UFA of 3
seems unjustified; a full value of 10 seems more
appropriate.

•	The third paragraph on pagel50 is confusing.
Evaporation from skin is assumed to not occur for
some of the dermal extrapolations, but it was
assumed to occur for others. Much more clarity is
needed. The last sentence of this paragraph makes
no sense. "Metabolism occurs in both oral and
dermal routes and inhalation is not as relevant to
dermal as absorption is more rapid by inhalation."
Also, many of the dermal extrapolations are
determined from inhalation data, which apparently
this sentence is saying is inappropriate.

•	In Section 5.2.4 Risk Characterizations the second
paragraph on page 150 is confusing. Is the critical
effect (respiratory metaplasia) evaluated by a
BMDL (lower confidence limit for BMD)? If so,
what is the point of raising concerns about using
LOAECs?

•	On what did EPA base the cutoff age of 16 for
adult that is used in several places? The Department
of Human Health Services (DHHS) standard cutoff
is age 18.

Public Comments:

•	The present draft evaluation confounds parameter
uncertainty and variability, when variability
(inherent natural variation) and uncertainty (lack of
knowledge) are distinct considerations in a

who may work with 1,4-dioxane or children ages 16 to 21
because EPA does not have information to indicate that
1,4-dioxane would preferentially affect women or
developing children."

•	In Section 2.4.1 and Table 4-3, EPA stated that for the
purpose of this assessment. EPA considered exposure of
the occupational users and non-users, which include but are
not limited to male and female workers of reproductive age
who are >16 years of age. Female workers of reproductive
age are >16 to less than 50 years old. Adolescents (>16 to
<21 years old) are a small part of this total workforce. The
occupational exposure assessment is applicable to and
covers the entire workforce who are exposed to 1,4-
dioxane.

•	EPA did not assess the toxicity of 1,4-dioxane for the
sediment environment. Because 1,4-dioxane is not expected
to sorb to sediment and will instead remain in pore water,
Daphnia magna and Gammarus pseudolimnaeus are two
species that feed through the entire water column and in
sediment were deemed to be an acceptable surrogate species
for sediment invertebrates. Therefore, EPA did not view
this as a data need. It has been well documented that D.
magna has been used to study the effects of hazardous
chemical to pore water and sediment contaminates such as
metals and organic compounds (Giesy etal., 1998; Othoudt
et al., 1991; Ristola etal., 1995; Coen and Janssen, 1998;
Suedel andRodgers 1996; CPA, 2004; Parkake^a/., 2010).
The results from these studies, procedures and protocols
have been used for establishing benchmark dose levels for
sediment toxicity.

Page 135 of 212

-------


probabilistic assessment. EPA needs to further
clarify parameter uncertainty and variability.

• EPA did not integrate information about
uncertainty and variability into an overall
characterization of the impact of uncertainty and
variability on estimated risks.



SACC,
24, 48,
58

Assumptions and Uncertainties - Exposure

SACC Recommendation: Provide more explanation

where requested.

•	In some scenarios where exposure is described as a
"reasonable worst case," the Evaluation does not
provide sufficient details on how exposure
estimates are determined to guess the percentile
represented by the exposure (e.g., is it an upper
10%, 5% or 1% scenario).

Public Comments:

•	With respect to its use of exposure models in
occupational settings, EPA should include
sensitivity analyses for models where assumptions
and uncertainty are prevalent.

•	Inaccurate assumptions used in dermal exposure
modeling may underestimate exposure. Issues
include: reliance on Bronaugh (1982); EPA's
extrapolation from inhalation to dermal risks
without considering flux dynamics that are
uniquely applicable to dermal absorption; EPA's
assumption that time equals infinity in its dermal
modeling (which overestimates evaporation and

•	EPA recognizes that some occupational exposure scenarios
rely on models or small datasets due to limited available
data but acknowledges that every available effort was made
to collect data. In preparation for the risk evaluation EPA
collected monitoring and exposure data directly from
industry, NIOSH, OSHA, KCNSC and other sources in
addition to an exhaustive search of published data.

•	As indicated earlier in this response to comment document,
the dermal exposure calculations have been updated and
validated by using sensitivity analysis and by comparing
test results reported by the researchers at the Kansas State
University, Manhattan, Kansas; and at the University
Erlangen-Nuremberg, Erlangen, Germany. The above-
mentioned references are cited in the revised risk evaluation
document (see Section 2.4.1.14 - Dermal Exposure
Assessment).

•	The revised risk evaluation document has been updated by
deleting Bronaugh (1982) and the relevant paragraph as the
Bronaugh (1982) cited data are not used in the dermal
exposure assessment. The dermal calculations have been
updated in the revised risk evaluation by including a
conceptual diagram (Figure 2-1), tiered analysis using: a)
updated calculations; b) sensitivity analysis to evaluate
chemical fluxes at various fractional absorption varying
from negligible (~0) to complete absorption (1.0) (see

Page 136 of 212

-------


underestimates absorption); mistakes in the dermal
dose equation calculations; and EPA's use of fixed
glove PFs of 5x, lOx, and 20x

Figure 2-2 ); c) overall comparison of modeled chemical
fluxes with in-vitro and ex-vivo test data reported in the
literature. Regarding the considerations of glove PF, the
EPA included a thorough analysis on incorporation of glove
protection, limitations and protections of glove use,
potential for occlusion in Appendices E (E5 and E6), and G.
Additional information are also included in Supplemental
document (Engineering Assessment of Occupational
Exposure for 1,4-Dioxane).

SACC,
24, 58,

Assumptions and Uncertainties - Hazard

SACC Recommendation: Provide more explanation

where requested.

•	On page 149, there is brief discussion of
metabolism. There is a seeming presumption that
metabolism is not required, but it is unclear what
the plausible mechanism would be for the parent
compound as opposed to a reactive intermediate.

•	In 5.3.3, the first full paragraph on page 149 states
that the main uncertainty for the human health
hazard is the unknown mode of action (MO A).
What is the basis for this assertion? Is this greater
than the potential for species difference or other
factors? The statement that there is no information
on the MOA for tissues other than the liver is not
strictly correct. The comprehensive evaluation of
the totality of the nasal toxicity/cancer data could
indeed provide insights into modes of action at this
site. Such an evaluation was not done.

•	In 5.3.3, on page 149 in the second paragraph it
states metabolic saturation is a proposed event in
the MOA. As highlighted above in this review, the
evidence for metabolic saturation is weak at best.

EPA has revised the discussion under the heading Human

Health Hazard Assumptions and Uncertainties in response to

these comments.

•	EPA agrees that there is uncertainty around whether
toxicity is caused by 1,4-dioxane itself or by metabolites
and describes this as a source of uncertainty.

•	EPA has modified the language on uncertainty related to
the MOA to state that it is "one source of uncertainty for
cancer risk estimates" rather than the main uncertainty.
EPA also deleted the statement that there is no information
about MOA for tissues other than liver. Instead, the final
language states that EPA concluded there was insufficient
information to support a specific MOA for any tumor type.

•	EPA considered the alternate explanation that the reduced
rate of metabolism is the result of perfusion limited
metabolism rather than metabolic saturation. Perfusion
limited metabolism would require liver toxicity that results
in reduced liver function to occur very rapidly in
toxicokinetic studies where metabolic saturation is
observed. In the MOA analysis, EPA notes that evidence in
support of metabolic saturation as a necessary key event
for liver tumor carcinogenesis is very limited.

•	EPA has inserted additional discussion of uncertainties
related to route-to-route extrapolation.

Page 137 of 212

-------
Indeed if 1,4-Dioxane is a high extraction
compound (in the liver) then perfusion limitation or
capacity limitation are potential events, not
metabolic saturation. Moreover, while hepatic
metabolic saturation may reduce active metabolite
formed in the liver it would also serve to diminish
hepatic clearance and increase delivery of 1,4-
Dioxane to other tissues.

Public Comments:

•	EPA should discuss the uncertainties that route-to-
route extrapolations introduce. The source EPA
cites for its approach to extrapolation (p. 150, citing
USEPA 2004) recommends that, at a minimum, a
thorough discussion of associated uncertainties be
included when such extrapolation is used. EPA
relied on oral-to-dermal extrapolation (p. 90) for
sub-chronic/chronic non-cancer outcomes, with
little acknowledgment of the uncertainties
associated with route-to-route extrapolation. The
inclusion of additional UF for route-to-route
extrapolation may be appropriate. As is
recommended for route-to-route extrapolation
generally and oral-to-dermal extrapolation
specifically, EPA should apply an additional
uncertainty factor of 10 to account for these
uncertainties.

•	EPA states in the Risk Determination section that
the agency's approach to estimating dermal
exposures "could overestimate risk, as EPA used
3.2%, the higher dermal absorption factor value
from Bronaugh (1982) in the risk evaluation.
Inclusion of an additional UF would affect the
conclusion that risk is overestimated.

• The format of the Unreasonable Risk Determination

section has been revised and no longer includes discussion
on these types of uncertainties. Uncertainties are now
discussed in Section 4.

Page 138 of 212

-------
SACC

Study Quality

SACC Recommendation: Each time Mattie etal.,

2012 is cited that EPA add "after Kasai et al., (2008)."

• Several Committee members were not comfortable
with the reliance on a single government report
that was not published in a peer-reviewed journal,
Mattie et al., (2012). It was also noted that Mattie
et al., (2012) is a limited repeat of the study
reported in Kasai et al., (2008) and that each time
Mattie et al., (2012) is cited Kasai et al., (2008)
should be included

Kasai et al., (2008) is a high-quality inhalation study, but it
only reports effects of a 13-week exposure. It did not evaluate
acute and short-term effects of 6-hour and two-week exposure
periods and cannot be used to inform dose-response for short-
term exposures. While some of the endpoints evaluated are
similar, the acute and short-term studies in Mattie etal., 2012
are not replications of the sub-chronic study reported in Kasai
et al., (2008). A previous two-week drinking water exposure
study performed by JBRC was not published and only limited
results of that study are available (See limited data available
from JBRC, 1998 at HERO ID: 196242). In addition, the
inhalation exposures evaluated in Mattie et al., are more
directly relevant for evaluating acute inhalation risk than
results from the drinking water study. While EPA recognizes
the weakness of the Mattie et al., (2012) report, it is a high-
quality study that provides some of the best available
information for acute effects of 1,4-dioxane. The original draft
risk evaluation misstated the results of the data quality
evaluation for the 2-week exposure study as medium quality
when the study in fact received a high data quality rating
(details of the scoring can be found in the supplemental file).
EPA has corrected this error. The POD derived from Mattie et
al., 2012 is supported by the broader weight of evidence,
including results of another two-week study on neurological
effects and by no effect levels reported in acute human
exposures.

59

Risk Calculations- Approach

Public Comment: Methods used in this risk
evaluation differ from existing evaluations of 1,4-
dioxane, mainly the occupational risk assessments
conducted by other US bodies (OSHA, ACGIH) and
international agencies (ECHA). EPA is encouraged to

EPA applied risk assessment methods tailored to the needs of
TSCA implementation. TSCA compels EPA to conduct risk
evaluations to determine whether a chemical substance
presents unreasonable risk, without consideration of cost or
other non-risk factors, under the conditions of use.
Occupational risk assessments and conclusions are performed
for a different purpose using a different set of assumptions and

Page 139 of 212

-------


re-examine the methodologies employed in those
occupational risk assessments and the conclusions.

considerations.

SACC

Risk Calculations- Error in Dermal Risk Estimates

SACC Recommendation: Correct the substantial
error in the dermal risk estimates that makes the risk
characterization invalid.

EPA made an error in dermal risk calculations presented in the
draft RE. By incorporating a dermal adjustment factor in both
the hazard and exposure portions of the risk calculations, EPA
had effectively compared PODs in terms of applied dermal
dose to predicted exposures in terms of absorbed dermal dose.
In the final RE, EPA has revised all dermal PODs to reflect
absorbed dermal dose rather than the applied dermal dose
calculated in the draft RE. This eliminates the error by putting
the exposure and hazard parts of the risk equation in common
terms that are more directly comparable. The revised risk
calculations predicted higher levels of risk from dermal
exposure than previously calculated.

SACC

Risk Calculations- Presentation and Clarity

SACC Recommendation: Provide more clarity where
requested. Specifically, add the suggested table to
clarify where EPA has and has not determined there to
be risk and unreasonable risk

•	One Committee member presented an idea for a
different table to clarify presentation of risk
estimates for a different audience.

•	Explain or give an example how values of column
1 (PF=1) in tables 5-9, 5-10, and 5-11 are
calculated. In table 5-10 the risk estimates for
import/packaging are given by ratios (e.g., 30/16
for PF=1). Why are they different from the other
bin 1 values?

•	Move 2nd paragraph page 143 to Section 3.4.1.14
where dermal exposure assessment is discussed.

EPA made the following changes in response to these

recommendations

•	EPA has included a new summary table showing risk
estimates for all COUs (with links back to corresponding
occupational exposure scenarios)

•	The risk estimates in all tables are calculated according to
the MOE and cancer risk equation described in Section
5.2.1. Exposure and hazard values used as the basis for all
risk calculations are available in the Risk Calculator
supplemental excel file. The risk estimates for
import/packaging in Table 5-10 in the draft RE were
indicating different risks for bottle vs. drum repackaging.
The bottle and drum repackaging scenarios have been split
into separate rows in the final risk evaluation to make this
clearer (in what is now Table 5-11)

•	The paragraph on dermal absorption (on p. 143 summarizes
information that is now included in Section 3.4.1.14 on
dermal exposure assessment and has been substantially

Page 140 of 212

-------




condensed.

19, 24,
47, 50,
51,54,
55, 56,
58

Potentially Exposed or Susceptible Subpopulations

Public Comments:

•	Failing to consider risks to vulnerable
subpopulations, such as children, the elderly,
pregnant women, and people who live near 1,4-
dioxane contaminated sites.

•	Risks to people living near disposal sites, including
(but not limited to) those living near so-called
"legacy" disposal sites were ignored.

•	There is specific concern for populations living in
Pleasantville, where underground tanks storing 1,4-
dioxane leaked, contaminating groundwater and
soil in adjacent areas.

•	Failing to consider exposures linked to disposal,
legacy uses, associated disposal, and legacy
disposal underestimates the background level of
exposures.

•	Environmental justice communities are often
disproportionately exposed and were not
considered.

•	The assumption that exposure to the general public,
including children, the elderly, and pregnant
women is adequately assessed, and that risks are
effectively managed by other statutes is not
supported.

•	No effort was made to identify all vulnerable
populations.

•	Communities where hydraulic fracturing is
common may be considered a sensitive
subpopulation.

EPA has determined that general population exposures due to
drinking water contamination, groundwater contamination, and
air emissions are under the jurisdiction of other statutes and are
outside the scope of this risk evaluation. Therefore,
subpopulations who may be exposed through these pathways
are outside the scope of the risk evaluation. However, the final
risk evaluation includes an evaluation of general population
exposures through recreational activities {i.e., swimming) in
ambient water bodies. See Section 1.4.2 of the final risk
evaluation.

EPA has added a footnote to the Executive Summary to clarify
that EPA did not identify any legacy uses of 1,4-dioxane. EPA
did not evaluate "legacy disposal" {i.e., disposals that have
already occurred) in the risk evaluation, because legacy
disposal is not a "condition of use" under Safer Chemicals, 943
F.3d 397.

Regarding body care and cosmetic products, they are excluded
from the definition of "chemical substance" per TSCA section
3(2) and are outside the scope of this risk evaluation.

EPA's review of the FracFocus reports on 1,4-dioxane indicates
that the 1,4-dioxane is likely present as an impurity in the
ethoxylated alcohols that are also named in the same reports.
EPA initially excluded production of 1,4-dioxane as a by-
product from certain other chemicals and presence as a
contaminant in industrial, commercial and consumer products
from the scope of the risk evaluation using EPA's discretion
under TSCA section 6(b)(4)(D). While EPA has addressed
some conditions of use related to 1,4-dioxane as a byproduct in

Page 141 of 212

-------
•	Consumers and adult women who use multiple
cosmetics and cleaning products may be considered
a susceptible subpopulation.

•	Adolescent girls use more body care and cosmetic
products than adult women and may be considered
a sensitive subpopulation for carcinogenic effects.

this risk evaluation, EPA expects that 1,4-dioxane exposures
associated with the use of ethoxylated alcohols used in
hydraulic fracturing fluids would be considered in the scope of
a risk evaluation for ethoxylated alcohols. In cases like this,
EPA believes its regulatory tools under TSCA section 6(a) are
better suited to addressing any unreasonable risks that might
arise from these activities through regulation of the activities
that generate 1,4-dioxane as an impurity or cause it to be
present as a contaminant than they are to addressing them
through direct regulation of 1,4-dioxane. This case-by-case
approach for byproducts exposures is consistent with the
various scenarios explained in the Risk Evaluation Rule, 82 FR
at 33730.

As discussed in Section 4.5 of the final risk evaluation, EPA did
not aggregate exposure across exposure routes (dermal,
inhalation or oral) for occupational, consumer, or general
population exposures. EPA chose not to employ simple
additivity of exposure pathways within a condition of use
because of the uncertainties present in the current exposure
estimation procedures. There is currently no PBPK model
available to facilitate evaluation of aggregate exposure from
simultaneous exposure through inhalation, dermal, and oral
contact with 1,4-dioxane. Without a PBPK model containing a
dermal compartment to account for toxicokinetic processes the
true internal dose for any given exposure cannot be determined,
and aggregating exposures by simply adding exposures from
multiple routes could inappropriately overestimate total
exposure. Conversely, not aggregating exposures in any manner
may potentially underestimate total exposure for a given
individual. EPA acknowledges in Section 4.3.2 that the decision
not to aggregate risk could result in an underestimate of risk.
This approach is consistent with the approach taken and peer

Page 142 of 212

-------




reviewed for the other solvent chemicals recently evaluated and
finalized.

SACC,
24, 56,
58

Potentially Susceptible Subpopulations

SACC Recommendation: EPA should more fully
address susceptible populations including pregnant
women or women who may become pregnant.
Modeling and sensitivity analysis may help address
these data gaps.

SACC Recommendation: Provide more clarity where
requested.

• How can the following conclusion be determined
given the absence of any separation of effects by
gender in these studies? "In developing the risk
evaluation, the EPA analyzed the reasonably
available information to ascertain whether some
human receptor groups may have greater exposure
than the general population to the hazard posed by
a chemical. The results of the available human
health data for all routes of exposure evaluated {i.e.,
dermal and inhalation) indicate that there is no
evidence of increased susceptibility for any single
group relative to the general population. Exposures
of 1,4-Dioxane would be expected to be higher
amongst workers and ONUs using 1,4-Dioxane as
compared to the general population." (page 151); In
fact, prior statements in the document contradict
this, stating: "Information on induction of liver
enzymes, genetic polymorphisms and gender
differences was inadequate to quantitatively assess
toxicokinetic or toxicodynamic differences in 1,4-

EPA has modified the language related to potentially exposed
or susceptible subpopulations in the risk characterization
section to be more transparent about uncertainties and data
gaps. The revised language acknowledges that some
subpopulations may be more susceptible due to lifestage,
genetic differences, pre-existing diseases, or other factors that
impair metabolism or increase susceptibility of the target
organs of 1,4-dioxane: ".. .it's possible that some
subpopulations are more biologically susceptible to the effects
of 1,4-dioxane due to genetic variability, pre-existing health
conditions, lifestage, pregnancy, or other factors that alter
metabolism or increase target organ susceptibility. For
example, people with liver disease may by more susceptible
due to reduced metabolism of 1,4-dioxane and increased
susceptibility of a target organ. EPA does not have sufficient
information about these potential sources of susceptibility to
quantitatively incorporate them into the risk assessment."

Page 143 of 212

-------


Dioxane hazard between animals and humans and
the potential variability in human susceptibility."

•	Page 151, last paragraph, does this describe an
approach that EPA will take in the future to
differentiate the risk for subpopulations with
varying exposure or severity of the health effects?
An alternative way to describe this is to discuss the
limitation of the current approach.

Public Comments:

•	General population and worker subpopulations that
have pre-existing conditions that affect the liver
may be considered a sensitive subpopulation.

•	There is significant evidence that the prenatal life
stage is more susceptible to carcinogens. Pregnant
women should be evaluated as a sensitive
subpopulation

•	The agency assumes that all workers are "healthy"
in its risk characterization (p. 132) but indicates
elsewhere that there may be numerous worker
subpopulations with pre-existing conditions that
affect the liver or other targets of 1,4-dioxane.

With regard to the comment on the last paragraph on page 151
of the draft risk evaluation, EPA has adjusted its approach to
unreasonable risk determinations for workers in the final risk
evaluation. While EPA has evaluated worker risk with and
without PPE, as a matter of policy, EPA does not believe it
should assume that workers are unprotected by PPE where
such PPE might be necessary to meet federal regulations,
unless it has evidence that workers are unprotected. In
consideration of these uncertainties and variabilities in PPE
usage, EPA uses the high-end exposure value when making its
unreasonable risk determination in order to address those
uncertainties.

SACC

Scope- Calculate risk for general population and
PESS

SACC Recommendation: Include estimates of risk to
general population and susceptible populations,
especially in other pathways of exposure such as
drinking water.

General population exposures via drinking water, ambient air,
and other pathways are not in the scope of this risk evaluation
(see Section 1.4.2). For example, EPA determined that
exposures to 1,4-dioxane through drinking water are under the
jurisdiction of the Safe Drinking Water Act. However, the final
risk evaluation includes an evaluation of general population
exposures through recreational activities {i.e., swimming) in
ambient water bodies. See Section 2.4.2 of the final risk
evaluation.

43, 46,

Scope- Regulatory Nexus

EPA found that exposures to the general population may occur

Page 144 of 212

-------
48, 58

Public Comments:

•	EPA explains the exclusion of consideration of
numerous exposure pathways, hazards, and
conditions of use based on the assumption that the
exposures evaluated in this risk evaluation are
"likely to represent the greatest areas of concern to
EPA" (p. 156). No support for this assertion is
provided.

a.	Reduced EPA enforcement provides less
assurance that exposures through the
excluded pathways are being effectively
managed.

b.	This whole-sub stance focus begins during
prioritization. The definitions of high- and
low-priority substances make clear it is the
"substance" that receives the designation,
not selected conditions of use, exposures, or
hazards.

•	When a pathway is excluded from further analysis,
EPA must have developed and applied a sound
basis for assessing the exposure level. EPA then
must consider how exposure from an individual
pathway combines with other sources of exposure.

•	EPA assumes that other environmental statutes
have or will sometime in the future address
potential risks of 1,4-dioxane that were not covered
in the current risk evaluation (e.g., general
population, consumer exposure, and PESS). EPA
cannot assume that other regulatory authorities
(e.g., OSHA, the Safe Drinking Water Act, the
Clean Water Act, DOT, and the Resource
Conservation Recovery Act) will adequately assess

from the conditions of use due to releases to air, water or land.
The exposures to the general population via drinking water,
ambient air and sediment pathways falls under the jurisdiction
of other environmental statutes administered by EPA, i.e.,
CAA, SDWA, and RCRA. As explained in more detail in
section 1.4.2, EPA believes it is both reasonable and prudent to
tailor TSCA risk evaluations when other EPA offices have
expertise and experience to address specific environmental
media, rather than attempt to evaluate and regulate potential
exposures and risks from those media under TSCA. EPA has
therefore tailored the scope of the risk evaluations for 1,4-
dioxane using authorities in TSCA sections 6(b) and 9(b)(1).
EPA did not evaluate hazards or exposures to the general
population via certain pathways in the risk evaluation, and as
such the unreasonable risk determinations for relevant
conditions of use do not account for exposures to the general
population for certain pathways. However, the final risk
evaluation includes an evaluation of general population
exposures through recreational activities (i.e., swimming) in
ambient water bodies. See Section 2.4.2 of the final risk
evaluation.

Page 145 of 212

-------


and manage risks from 1,4-dioxane. OSHA
standards do not protect public sector workers or
construction workers who are classified as
independent contractors and are therefore not
covered by OSHA. EPA's risk evaluations indicate
that such contractors would be exposed to 1,4-
dioxane, yet EPA does not discuss how their status
as workers affects EPA's PPE analyses and
assumptions.



SACC,
47, 55,
56

Aggregate Exposures

SACC Recommendation: EPA should evaluate
combined exposures through several pathways,
including pathways that were not evaluated such as
drinking water.

• Page 152, line 16, states: "As a result of the limited
nature of all routes of exposure to individuals
resulting from the conditions of use of 1, 4-
Dioxane, a consideration of aggregate exposures of
1, 4-Dioxane was deemed not to be applicable for
this risk evaluation": This does not seem a strong
justification for not considering aggregated
exposure. If all routes of exposure are of limited
nature, then would not a single route be of even
more limited nature? From a practical standard
point, there are technical challenges to combining
cancer and non-cancer risk when different
approaches are taken to quantify non-cancer risk
(MOE) and cancer risk (slope factor). While it is in
principle possible to combine different health
outcomes or risk metrics into a joint estimate of
risk associated with aggregated exposure from

TSCA section 6(b)(4)(F)(ii) directs EPA to "describe whether
aggregate or sentinel exposures to a chemical substance under
the conditions of use were considered and the basis for that
consideration" in risk evaluations. EPA defines aggregate
exposures as the combined exposures to an individual from a
single chemical substance across multiple routes {i.e., dermal,
inhalation, or oral) and across multiple pathways {i.e., exposure
from different sources). 40 CFR 702.33. EPA has determined
that using the high-end risk estimate for inhalation and dermal
risks separately as the basis for the unreasonable risk
determination is a best available science approach. There is low
confidence in the result of aggregating the dermal and
inhalation exposures and risks for this chemical if EPA uses an
additive approach, due to the uncertainty in the data. EPA does
not have data that could be reliably modeled into the aggregate
exposure, such as would occur with a PBPK model. Using an
additive approach to aggregate exposure and risk in this case
would result in an overestimate of risk. Given all the limitations
that exist with the data, EPA's approach is the best available
science.

As explained in the scope document, 1,4-dioxane may be found
as a contaminant in consumer products that are readily available
for public purchase. In the final risk evaluation, eight consumer

Page 146 of 212

-------


multiple route, this is an area that requires more
research. EPA can be more transparent with these
limitations.

Public Comments:

• Overall, there is concern that cancer risk was

underestimated because a) the totality of exposures
(e.g., consumer uses and drinking water) were not
considered and b) risks from different routes of
exposure (e.g., dermal and inhalation) were not
combined and, therefore, potential additive effects
were not assessed. This is of particular concern
since the IRIS database lists 1,4-dioxane as a
probable human carcinogen.

conditions of use are evaluated based on the uses identified in
EPA's 2015 TSCA Work Plan Chemical Problem Formulation
and Initial Assessment of 1,4-Dioxane ( ). See
Section 2.4.3 of the final risk evaluation.

SACC,
21, 22,
46,

PPE Assumptions

SACC Recommendation: Further explain PPE use
and its relation to risk assessment and risk evaluation.
• In many incidences EPA shows the use of

protective devices with a certain level of Assigned
Protection Factor (APF) would bring the risk level
(MOE or cancer slope factor) to a reasonable level
(in reference to benchmark). If the intention is to
demonstrate the effectiveness of use of a protective
device, EPA should report the results associated
with the smallest Protection Factor (PF) that
achieves the desired level of risk reduction when
such a device is available. The Evaluation's
presentation, in the current version, is inconsistent
in that in some cases the "risk level" (MOE or
cancer slope) is below and in other cases remained

The OSHA regulations at 29 CFR 1910.132 require employers
to assess a workplace to determine if hazards are present or
likely to be present which necessitate the use of personal
protective equipment (PPE). If the employer determines
hazards are present or likely to be present, the employer must
select the types of PPE that will protect against the identified
hazards, require employees to use that PPE, communicate the
selection decisions to each affected employee, and select PPE
that properly fits each affected employee. OSHA has
established a permissible exposure limit (PEL) of 100 ppm (8-
hour TWA) for 1,4-dioxane. However, as noted on OSHA's
website, "OSHA recognizes that many of its permissible
exposure limits (PELs) are outdated and inadequate for
ensuring protection of worker health. Most of OSHA 's PELs
were issued shortly after adoption of the Occupational Safety
and Health (OSH) Act in 1970, and have not been updated
since that time." OSHA provides an annotated list of PELs on
its website, including alternate exposure levels. For 1,4-

Page 147 of 212

-------
above the benchmark level.

Public Comments:

•	In the risk evaluation, EPA determines that
occupational inhalation of 1,4-dioxane causes risks
up to 48 times greater than levels EPA deems
acceptable; however, EPA concludes that few risks
associated with 1,4-dioxane are unreasonable. EPA
assumes that exposed workers would be provided
with, would consistently use, and would be
adequately protected by personal protective
equipment ("PPE"). EPA should not assume use
of PPE in the risk evaluation to avoid concluding
"unreasonable risk".

•	It is not clear how EPA has considered compliance
with OSHA's worker protection standards in the
1,4 dioxane risk evaluation. This should be
clarified in the document.

•	EPA cites no evidence that workers exposed to 1,4-
dioxane are provided with, or consistently use,
PPE. The studies that EPA relies on to determine
occupational exposure levels do not mention PPE
use.

•	EPA discounts the reported exposure levels based
on the assumption that all workers are provided
with, and use, appropriate PPE.

•	EPA's risk evaluation assumes the use of PPE and
also the specific types and protectiveness of such
equipment. For 1,4-dioxane, EPA assumes that
workers will wear respirators with an average
assigned protection factor ("APF") of at least 50
and impervious gloves with a protectiveness factor
("PF") of at least 20. The Safety Data Sheets
("SDS") referenced by EPA—described below, are

dioxane, the alternates provided are the California OSHA PEL
of 0.28 ppm and the ACGIH TLV of 20 ppm.
(https://www.osha.gov/dsg/annotated-pels/tablez-l .html) EPA's
approach for developing exposure assessments for workers is to
use the reasonably available information and expert judgment.
When appropriate, in the risk evaluation, EPA will use
exposure scenarios both with and without engineering controls
and/or personal protective equipment (PPE) that may be
applicable to particular worker tasks on a case-specific basis for
a given chemical. Again, while EPA has evaluated worker risk
with and without PPE, as a matter of policy, EPA does not
believe it should assume that workers are unprotected by PPE
where such PPE might be necessary to meet federal regulations,
unless it has evidence that workers are unprotected. For the
purpose of determining whether or not a condition of use
presents unreasonable risks, EPA incorporates assumptions
regarding PPE use based on information and judgment
underlying the exposure scenarios. These assumptions are
described in the unreasonable risk determination for each
condition of use, in section 5.2 of the risk evaluation.
Additionally, in consideration of the uncertainties and
variabilities in PPE usage, EPA uses the high-end exposure
value when making its unreasonable risk determination in order
to address those uncertainties. EPA has also outlined its PPE
assumptions in section 5.1 of the risk evaluation. Further, in the
final risk evaluation for 1,4-dioxane, EPA has determined that
most conditions of use pose an unreasonable risk to workers
even with the assumed PPE.

Page 148 of 212

-------


not binding on employers. According to EPA, "a
specific glove material or protection factor rating
was not provided" in the 1,4-dioxane SDSs. An
SDS for a chemical reagent which is 50-60% 1,4-
dioxane by weight states that "the use of this
product should not require respiratory protection."
Other SDSs referenced in the draft risk evaluation
recommend unspecified respirators only "if
workplace exposure limit(s) of product or any
component is exceeded."



SACC,
56

Non-cancer Benchmark MOEs

SACC Recommendation: Provide more clarity where

requested:

•	Define Benchmark MOE formally, preferably using
an equation, giving adequate references and
interpretation.

Public Comments:

•	EPA should use a unified linear approach for dose-
response analysis and risk calculations for all
carcinogens and non-carcinogens as recommended
by the National Academy of Sciences (NAS); EPA
should not use the margin of exposure (MOE)
approach.

•	Use of MOEs is a restrictive approach. MOEs do
not provide a risk estimate, but a single number
similar to a reference dose. Similar to cancer, a
non-MOE approach should be applied to risk
calculations for all other health endpoints.

The MOE is a standard risk assessment approach. EPA applies
an MOE approach because it allows for the presentation of a
range of risk estimates and does not create a "bright line" for
regulation.

EPA has inserted a definition of the benchmark MOE in the
"human health risk estimation approach".

SACC,
24, 50,
54, 58

Cancer Benchmarks

SACC Recommendation: Provide more clarity where
requested:

ONUs work in the industrial and commercial environment. As
such, consistent with 2017 NIOSH guidance, EPA used lxlO"4
as the benchmark for the purposes of the unreasonable risk

Page 149 of 212

-------
•	Explain why 10-4 is appropriate for ONUs given
that 10-6 is usually used for the general population.

•	Give reference for the cancer benchmark level of
10-4 (e.g., Table 5-3). Referencing the key findings
in the result tables would improve the presentation.

Public Comments:

•	EPA relied on NIOSH guidance in order to
establish 1 x 10-4 as the cancer risk benchmark for
workers, although acknowledging that other laws
have standards that differ from TSCA's (p. 153).

•	It is unclear why a cancer risk benchmark of one-
in-10,000 was used as the threshold for
unreasonable risk to workers who inhale but "do
not typically directly handle" 1,4-dioxane in the
scope of their jobs.

•	EPA guidelines state that the appropriate cancer
risk threshold can be anywhere from one-in-10,000
to one-in-one million. Why was the less protective
one-in-10,000 threshold chosen?

•	In this risk evaluation, EPA has set a risk level for
the entire worker population that is the same as the
level EPA elsewhere set for the most exposed
individual in a population. EPA then invokes this
level repeatedly to find the majority of conditions
of use of 1,4-dioxane to pose no risk to any
workers, thereby subjecting many tens of thousands
of workers to cancer risks that are as much as two
orders of magnitude higher than warranted. This
approach must be rejected on scientific as well as
legal grounds.

determination for individuals in industrial and commercial work
environments who are exposed to 1,4-dioxane.

As noted in the draft risk evaluation (Section 5.1.1), EPA relied
on NIOSH guidance (Whittaker et al, 2016) when choosing the
10"4 cancer risk benchmark to evaluate risks to workers from
1,4-dioxane exposure. NIOSH's mandate, on pg iii of Whittaker
eial, (20161 is to: "... describe exposure levels that are safe
for various periods of employment, including but not limited to
exposure levels at which no employee will suffer impaired
health or functional capacities or diminished life expectancy as
a result of his work experience." Although NIOSH guidance, p.
20, states that: "exposures should be kept below a risk level of 1
in 10,000, if practical [emphasis added]" EPA adheres to the 1
in 10,000 benchmark during the risk evaluation stage for TSCA
chemicals.

Other precedents (e.g., Office of Water; Office of Air) are the
basis for cancer benchmarks to be used for risks to the general
population. Consistent with these precedents, EPA applied a
cancer risk benchmark of 10"6 to evaluate cancer risks for
consumers.

Standard cancer benchmarks used by EPA and other regulatory
agencies are an increased cancer risk above benchmarks
ranging from 1 in 1,000,000 to 1 in 10,000 (i.e., lxlO"6 to 1x10"
4) depending on the subpopulation exposed.

EPA, consistent with 2017 NIOSH guidance, used lxlO"4 as the
benchmark for the purposes of this unreasonable risk
determination for individuals in industrial and commercial work
environments. It is important to note that cancer risk thresholds
(lxlO"4 or lxlO"6) are not a bright line and EPA has discretion
to make unreasonable risk determinations based on other

Page 150 of 212

-------


•	For all 10 COUs, inhalation cancer risk levels for
workers are above 1 in 100,000 - even with
respirator use - for both central tendency and high-
end exposures. For 7 of the 10 COUs, high-end
cancer risk levels for workers are above 1 in
10,000. For 1 of these, even an APF=50 respirator
is not sufficient to get the high-end cancer risk
below this risk level. For the other 6 COUs,
respirators are necessary to get the high-end cancer
risk levels below 1 in 10,000 (APF=50 for 2 COUs;
APF=25 for 3 COUs; APF=10 for 1 COU). For 5 of
10 COUs, central tendency cancer risk levels are
also above 1 in 10,000. For these, respirators are
necessary to get the central tendency cancer risks
risk levels below 1 in 10,000 (APF=25 for 1 COU;
APF=10 for 4 COUs).

•	For Dermal cancers, For 9 of 11 COUs, dermal
cancer risk levels for workers are above 1 in
100,000 - even with PF=20 glove use. For 9 of 11
COUs, dermal cancer risk levels for workers are
above 1 in 10,000 - even with PF=5 glove use. For
8 of these, PF=10 gloves still leave risk above 1 in
10,000, and For 6 of these, even PF=20 gloves are
not sufficient to get risk below 1 in 10,000.

•	Studies with statistically significant data, in regard
to human health, were dismissed, which also
contradicts EPA cancer guidelines

benchmarks as appropriate. See section 5.1.1.2 of the risk
evaluation for additional information.

For the purposes of determining whether or not an occupational
condition of use presents unreasonable risks, EPA incorporates
assumptions regarding PPE use based on information and
judgement underlying the exposure scenarios. These
assumptions are described in the unreasonable risk
determination for each condition of use, in section 5.2.
Additionally, in consideration of the uncertainties and
variabilities in PPE usage, EPA uses the high-end exposure
value when making its unreasonable risk determination in order
to address those uncertainties. EPA has also outlined its PPE
assumptions in section 5.1. Further, in the final risk evaluation
for 1,4-dioxane, EPA has determined that most conditions of
use pose an unreasonable risk to workers even with the
expected PPE.

SACC

Risk Determination- Clarity

SACC Recommendation: Provide more clarity where

The format of the unreasonable risk determination section has
been revised, which should address the comments in bullets 1,
2, and 4.

Page 151 of 212

-------
requested:

•	Under Risk Considerations, EPA acknowledges
that some models and/or assumptions could lead to
overestimation of risk (conservative assumptions)
and then the text proceeds to give examples of
where these uncertainties might exist. However,
the text proceeds to read "For this pathway, EPA
expects that the risks are not underestimated."
Shouldn't this read: "As a result, EPA expects that
the risks are not underestimated for this pathway."
given this seems a likely conclusion if models and
assumptions increase the likelihood of
overestimation?

•	In Section 6, related to the risk characterization
assumptions, the text of the table (first on page 158
and then repeated) indicates that whole body
inhalation studies overestimate the risk for portal
of entry effects because of uncertainty with respect
to the actual doses received. The justification for
why an overexposure would be expected from
consideration of only whole-body inhalation
exposures is not clear. The risk evaluation should
discuss more the pathways (indirect via blood
concentration or direct via direct deposition via
dust inhalation and/or preening) expected for nasal
tissue exposures and whole-body inhalation study
data would lead to overestimation of risk.

•	On page 173, for disposal, workers are
inappropriately included in both assessment
statements. There is either an unreasonable risk of
injury or not, it can't be both: "Presents an

For bullet 3, EPA thanks the SACC and has corrected the error.
Workers should not appear in both places.

Page 152 of 212

-------


unreasonable risk of injury to health (workers)";
"Does not present an unreasonable risk of injury to
health (workers and occupational non-users) or to
aquatic vertebrates, aquatic invertebrates, and
aquatic plants from exposures to 1,4-Dioxane in
surface waters.

• In the summary table (Table 6-1) that provides
EPA's risk determinations, each "Risk
Considerations" section should discuss factors that
may over- or under-estimate the risk.



SACC,
19, 24,
46, 48,
58

Risk Determination- Unreasonable Risk
Conclusions

SACC Recommendation: Modify statements of no
unreasonable risk with appropriate qualifiers such as
"if appropriate personal protective equipment is used"
or if "engineering controls are used."

SACC Recommendation: Provide more clarity where
requested:

•	One member noted that in Table 6-1 there are
instances where the MOE is less than 30 with PPE,
but no unreasonable risk is noted and suggested this
is an error. This points to the difficulty interpreting
these tables as other Committee members noted the
determination of unreasonable risk came from a
different table.

Public Comments:

•	EPA finds no unreasonable risk for some
conditions of use despite having estimated MOEs

In response to the comments received, the format of the
unreasonable risk determination section has been revised and
there is increased clarification regarding when assumptions
were made regarding use of PPE.

The revised format of the unreasonable risk determination
section also reduces the challenges identified with table
interpretations comment.

The format of the unreasonable risk determination section has
been revised, which should address this comment.

With respect to the risk determinations discussed by the public
commenters, many of the risk estimates have been revised for
the final risk evaluation and the unreasonable risk
determinations have changed for some conditions of use. In the
final risk evaluation, all but three of the conditions of use

Page 153 of 212

-------
only slightly above the benchmark MOE.
o Table 5-5, rightmost column for Film
Cement: calculated MOE of 31 vs
benchmark MOE of 30 is deemed not to
represent unreasonable risk,
o Table 5-4, rightmost column for Industrial
Use: calculated MOE of 338 vs benchmark
MOE of 300 is deemed not to represent
unreasonable risk.

EPA has not explained how the MOEs being in
"proximity" to the benchmark negates the finding
of unreasonable risk. While EPA emphasizes that
some uncertainties might overestimate the risk
presented by these conditions of use, EPA fails to
account for how these or other uncertainties might
underestimate the risk.

EPA finds no unreasonable risk even when the
high-end risk exceeds relevant benchmarks, an
approach that is not adequately protective.
EPA calculates central tendency and high-end
exposure scenarios for all occupational uses.

Where the high-end scenario does not result in
findings of unreasonable risk (assuming the use of
PPE), EPA relies on that scenario for its risk
determinations.

In at least two instances, however, EPA discounts
high-end scenarios where the calculating margin of
exposure fell outside EPA's acceptable range and
relied on central tendency assumptions in order to
avoid finding unreasonable risk. EPA offers no
rationale for this approach.

The central tendency scenario is not protective of

present unreasonable risks to workers, ONUs, or both,
including Film Cement and Industrial Use.

For the purposes of determining whether or not a condition of
use presents unreasonable risks, EPA incorporates assumptions
regarding PPE use based on information and judgement
underlying the exposure scenarios. These assumptions are
described in the unreasonable risk determination for each
condition of use, in section 5.2. Additionally, in consideration
of the uncertainties and variabilities in PPE usage, EPA uses the
high-end exposure value when making its unreasonable risk
determination in order to address those uncertainties. EPA has
also outlined its PPE assumptions in section 5.1.

While EPA believes that discussions of the rationale for the
determination of unreasonable risk is outside the scope of the
SACC, EPA is committed to providing the public with
sufficient information on the basis for that determination.

TSCA requires EPA to determine whether chemicals in the
marketplace present unreasonable risks to health or the
environment. While the law does not specifically define this
term, during the risk evaluation process EPA weighs a variety
of factors including the health effects of the chemical on
humans or the environment, who are exposed (including any
sensitive subpopulations), the severity of the hazard, and
uncertainties. This approach is outlined in EPA's 2017
Procedures for Chemical Risk Evaluation Under the Amended
Toxic Substances Control Act rule (risk evaluation rule)
preamble on how risk evaluations will be conducted. [82 FR
33726, at 33735 (July 20, 2017)]

EPA has provided more information on uncertainties in Section
4.3.

Page 154 of 212

-------
workers. EPA assumes a 31-year career (nine fewer
than the OSHA default level). In its risk evaluation,
EPA must also evaluate and account for the risks
posed by intermittent peak exposures.

•	In this risk evaluation, EPA is more likely to
determine unreasonable risk exists for workers
where risks greater than the acceptable benchmarks
are identified for both central tendency and high-
end exposures under the conditions of use.
However, worker exposure to contaminants in
drinking water or other "regulated pathways" under
central tendency or high-end conditions is not
evaluated, so worker exposures are being
underestimated under both scenarios.

•	The SACC should evaluate whether EPA has
sufficiently established the scientific basis upon
which to determine that EPA's risk determinations
are supported by the risk characterization. If the
SACC determines that EPA has not established that
scientific basis, the SACC should provide
suggestions as to how EPA can improve those
components of the risk evaluation.

•	EPA's failure to address uncertainty in its
quantitative risk characterizations and
determinations means that it cannot conclude that
1,4-dioxane does not present unreasonable risks
under its conditions of use.

•	EPA states that the degree of confidence or
uncertainty in the data it has will be a factor in
making its risk determinations (pp. 152, 154), but
never explains how this will factor in.

Page 155 of 212

-------
7. Editorial Comments

Kdilorisil ('oiniiicnls

22, 24,
58

•	The draft risk evaluation is missing citations, sources, and tables, and
cites sources that are not publicly available, contain no HERO entries,
lack hyperlinks, and/or cannot be located through internet searches.
These include:

•	Bronaugh (1982), which EPA cites as the basis for its skin
absorption estimates, is a chapter in a book that cannot be located.

•	USEPA (2018a), which EPA used to estimate exposures to 1,4-
dioxane in spray foam applications in the absence of monitoring
data (p. 68).

•	McConnell (2013), a technical report which EPA uses to describe
cytotoxicity as a potential MOA of liver toxicity and cancer.

•	BASF (2016) and BASF (2018). The former includes no link in
HERO. The latter includes a link, which is routed to an "error" in
regulations.gov.

•	JBRC (1998), a 2-year animal study conducted in Japan.

•	Dennerlein, K. etal., Studies on percutaneous penetration of
chemicals - Impact of storage conditions for excised human skin,
Toxicology in Vitro, Volume 27, Issue 2, 2013, pp. 708-713.

•	Communications with the racing authorities banning the use of 1,4-
dioxane as fuel or fuel additive do not appear to be publicly
available for consideration by the public.

•	A table explaining the calculations for section 4.2.6.2.5: Chronic Non-
Cancer POD for Dermal Exposures extrapolated from Chronic
Inhalation Studies (p. 117) should be included to ensure transparency.

•	Tables presenting dermal data need revision for accuracy and
clarification:

•	Table 5-9:

•	EPA included missing citation sources
and tables, and corrected errors in the
final risk evaluation as noted below:

•	The Bronaugh (1982) book-chapter is
available at several libraries in the
Washington DC area, including the
Library of Congress.

•	EPA has added a link to the USEPA
(2018a) HERO entry (HERO ID:
5080424) to provide public access to the
referenced Generic Scenario for the
Application of Spray Polyurethane
Foam Insulation.

•	McConnell (2013) and JBRC (1998) are
unpublished reports that are not publicly
available. EPA has therefore posted
them to the docket to ensure public
access.

•	The Dennerlein et al., study is a
published study that is publicly
available online. EPA cannot post this
copyrighted material.

•	(Part 1): BASF (2016), with HERO ID:
5079874, has been fixed in HERO and
now links to the correct document.

Page 156 of 212

-------
i.	Columns showing the results for central tendency scenarios
need to be added.

ii.	The values in the PF=5 column that are <3400 need to be
boldfaced.

•	Tables 5-10 and 5-11

i. Columns showing the results for central tendency as well as
high-end scenarios need to be added.

EPA's calculated acute COC is inconsistently reported. In the text, it is
listed as 59,800 ppb (p. 35), while in Appendix C it is listed as 20,000
ppb (p. 70).

Table 6-1 (starting on page 157) requires more detailed information and
editing.

•	Other than the "risk estimate" statements, (which include the types
of PPE used and the Tables in the document that support these),
there are no specific citations to the pages or tables in the risk
evaluation that support EPA's conclusions about "unreasonable
risk driver," "driver benchmarks," the "systematic review"
findings or the "risk considerations." These citations should be
added.

•	EPA should consider including a modified table that represents the
relevant endpoints and drivers, potentially color-coded with regard
to those that exceed benchmarks.

•	The subheadings in the table for each of the conditions of use are
not identical throughout the table. Does this imply differences?

•	The consideration of PPE in this table is confusing; in the
conditions of use section, the reader is left to assume workers are
not being protected with PPE, which then contradicts statements
made in the "risk considerations" section.

Section 6.2 of the draft requires more attention if it is going to serve as
a solid risk communication tool to the public. For example, Section 6.2
summarizes the risk determination in 1.5 pages but does not include a
sub-header for "workers," even though workers are a primary focus of
the draft risk determination.

(Part 2): (BASF, 2018a), with HERO
ID: 5079871 provides a URL to a file in
the 1,4-dioxane docket, but the link had
a typo. EPA corrected the link in HERO
and resolves this comment.

EPA updated dermal tables 5-9, 5-10,
and 5-11 to include central tendency and
resolved various boldfaced formatting
errors for PF=5. Note, EPA reads the
comment that "values in the PF=5
column that are <3400 need to be
boldfaced" as a typo that should read
<300, the Benchmark MOE.
EPA inadvertently calculated the short
term (acute) concentrations of concerns
(COCs) for fish rather than for the algal
endpoint. EPA has since updated the
COC for the correct endpoint. In
addition, EPA has updated the
calculations used for determining the
assessment factor for the acute endpoint.
EPA appreciates the feedback that the
unreasonable risk determination section
and table should include additional
detail and have greater consistency
across chemicals. EPA has revised the
formatting and clarity of the
unreasonable risk determination section.
In addition, upon issuance of the risk
evaluation (a scientific and policy
document), EPA intends to provide risk

Page 157 of 212

-------
EPA's HERO entries for references in the draft risk evaluation that
cite the REACH dossier for 1,4-dioxane erroneously state that the
author of the study summaries and underlying studies is ECHA itself,
when in fact it is the registrant. Examples of such erroneous entries
include:

•	https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/
3809089 (linked to on p. 211 of the draft risk evaluation as
"ECHA, 1996")

•	https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/
3809095 (linked to on p. 223 of the draft risk evaluation as
"ECHA, 2014").

There is a discrepancy in the "disposal" condition of use (p. 173),
which both declares: "Presents an unreasonable risk of injury to health
(workers)" - and in the next paragraph reads "Does not present an
unreasonable risk of injury to health (workers and occupational non-
users).	

communication materials applicable to a
broad range of stakeholders.

This is an error. The final unreasonable
risk determination correctly reflects that
the disposal condition of use presents an
unreasonable risk to workers and ONUs.

8. Supplemental Analysis

Public Comments

#

Summsirv of C omments lor Specific
Issues Kchitcd to (lie Supplcincntnl Ansilvsis

I.PA/OPPT Response

Peer Review & Public C omment Intension Requests

77, 82,
83, 85,
88, 89

•	EPA has not provided sufficient time for the public to review and
comment on the Supplemental Analysis, which contains new and highly
technical analyses of risks that were not addressed in EPA's prior risk
evaluation for 1,4-dioxane. EPA provided only 20 days to comment on
its new evaluation of 1,4-dioxane (one of which was Thanksgiving),
despite EPA regulations mandating a 60-day comment period on TSCA
risk evaluations.

•	EPA granted a 30-day extension of the public comment period for the
Pigment Violet 29 supplemental draft, 10 days longer than the period it

• As EPA explained in the Draft
Supplemental Analysis to the Draft
Risk Evaluation, the draft
supplemental analysis was not peer
reviewed for the sake of expediency
to finalize the first ten risk
evaluations.TSCA section 6(b)(4)(H)
requires a 30-day notice and
comment period on the draft risk

Page 158 of 212

-------


has now afforded for 1,4-dioxane. PV-29 has narrower uses and impacts
a far smaller population than 1,4-dioxane.

• This truncated process is without precedent. It has curtailed the public's
ability to provide informed feedback on the supplement.

evaluation prior to publication of the
final risk evaluation. Additionally, 40
CFR 702.49(a) provides for a 60-day
public comment period. EPA
complied with these statutory and
regulatory requirements by providing
a 60-day comment period from July
1, 2019 to August 30, 2019 on the
draft risk evaluation.

•

•	TSCA section 6(b)(4)(H) requires a
30-day notice and comment period on
the draft risk evaluation prior to
publication of the final risk
evaluation. Additionally, 40 CFR
702.49(a) provides for a 60-day
public comment period. EPA
complied with these statutory and
regulatory requirements by providing
a 60-day comment period from July

1, 2019 to August 30, 2019 on the
draft risk evaluation.

•	EPA is working diligently to publish
the final risk evaluation on 1,4-
dioxane as soon as possible. Given
the relatively small number of
conditions of use addressed by the
supplemental analysis, EPA believes
that 20 days, while expedited, was
sufficient to allow for public
comment. This will enable EPA to
quickly consider the public

77, 82,
83, 85,
88, 89

•	EPA's decision to dispense with peer review for the supplemental
evaluation is in contrast to PV-29, irresponsible, and compromises the
credibility of the Agency. Now that EPA has broadened the scope of
the evaluation to include exposures affecting a broad segment of the US
population, further peer review is essential to assure protection of public
health.

•	The supplement provides entirely new analyses related to two topics
that EPA excluded from its prior evaluation: (1) the risks associated
with the presence of 1,4-dioxane in consumer products and (2) the risks
associated with exposure to 1,4-dioxane from swimming and fishing in
contaminated water. The "Supplemental Analysis" is the first and only
evaluation of those risks prepared by EPA, and it must be treated as a
risk evaluation for the purpose of public comment.

•	Neither TSCA nor EPA's Risk Evaluation Rule provides an exception
from peer review of risk evaluations based on "expediency." EPA made
clear that issuance of a partial risk evaluation - as it has done in the
Supplement - still requires that the document be subject to peer review
- which it has not done. It also makes clear that a risk determination
cannot be made on a condition of use without peer review.

•	EPA must integrate its new analysis into an integrated risk evaluation
and subject that full document to peer review before finalizing the risk
evaluation and risk determinations. In addition, the public must be
afforded the opportunity to review and comment on that integrated
document, in-depth review of the supplement by the public and
independent scientists is essential to assure that the final evaluation is
credible and compliant with TSCA.

Page 159 of 212

-------




comments, revise the document as
appropriate, and publish the final risk
evaluation without further delay.

Svslcinsilic Ucviow/Lilomlurc Son re h mid Screening Comments

N.">

•	Because EPA is ignoring contaminated drinking water and exposure via
ingestion, EPA's literature search excludes key sources of exposure
information relevant and necessary to comprehensively characterize
actual risks to relevant receptors from the presence of 1,4-dioxane as a
byproduct in consumer and commercial products. For example, Figure
1-3 indicates that 13,296 of 21,373 references were excluded as non-
consumer references. While some of these may not be relevant to fully
assessing risk from 1,4-dioxane as a byproduct, others may well be
relevant to characterizing aspects of risk EPA has ignored.

•	EPA's supplemental literature selection process commenced with a
search for references specific to assessing 1,4-dioxane exposure to
consumers, yielding 8,077 filtered references. EPA then applied a
machine learning model to rank "how similar the filtered references
were to a pre-determined set of consumer references (positive seeds),
and how unsimilar [sic] the filtered references were to a pre-determined
set of non-consumer references (negative seeds)." (p. 13) EPA does not
appear to have provided the list of the positive and negative seeds in the
risk evaluation, the Supplement, or the supporting documents; it needs
to do so. Additionally, EPA should explain the basis for its choice of
relevancy cut-offs (0.1 for references with just titles and 0.4 for
references with abstracts) and the decision to use different relevancy
cut-offs dependent on the presence or absence of an abstract; the latter
seems especially arbitrary and potentially biased against sources of
information that are not structured in the conventional peer-reviewed
literature format. EPA's reliance on the machine learning model is
clearly consequential as it reduced the number of supplemental
information sources from 8,077 to 239. Additional explanation of key
filtering/cut-off decisions by EPA is necessary if there is to be any

•	The systematic review effort was
tailored to the scope of the risk
evaluation. Therefore, during
screening steps, literature that is not
relevant to the evaluated pathways
were not included for further
evaluation, extraction, and
integration.

•	EPA has updated its supplemental
file [Consumer References, Data
Screening] to include a list of the
positive and negative seeds utilized in
the referenced machine learning
process. The relevancy cut-off
language has been updates to reflect
the process accurately. EPA utilized a
0.1 relevancy cut-off for all sources.

•	The supplemental search was
initiated on a chemical name basis.
Therefore, the pool of literature was
very large relative to sources that
contained useful information on
consumer exposures, consumer
products, emission rates from
consumer products, etc. The process
is described and shown in Section
1.5.1 of the final risk evaluation.

Page 160 of 212

-------


confidence in the decisions that excluded so many sources from full
consideration.

• Of the 545 references screened (239 supplemental references plus 272
initial references from the risk evaluation), only 37 were subject to
formal data evaluation, in addition to 17 sources that were qualitatively
evaluated. EPA does not but needs to provide some explanation for why
the vast majoritv of studies it screened were not subject to evaluation



(iciicml Population Kxposurc Comments

78

• EPA should also evaluate the risks of consuming contaminated fish
because the supplemental analysis documents a bioconcentration factor
(BCF) of 0.9, resulting in tissue levels nearly equivalent to the water
concentration. Even though 1,4-dioxane is not bioaccumulative, fish
tissue concentrations may still pose a risk to consumers.

• EPA has determined that fish
consumption does not present an
unreasonable risk to the general
population. As described in Section
2.4.2 of the risk evaluation, 1,4-
dioxane's bioaccumulation factor
(BAF) indicates that concentrations
in fish tissues are expected to be
lower than aqueous concentrations
and supports the expectation that fish
ingestion is not a primary pathway of
human exposure for 1,4-dioxane.
Given its hydrophilic properties and
short half-life, 1,4-dioxane is not
expected to accumulate in tissue.

78, 89

• EPA states that risks of surface water contaminated with 1,4-dioxane on
swimmers were evaluated owing to the lack of human health water
quality criteria and thus regulation under the Clean Water Act. EPA
inaccurately indicates that human health criteria are designed to protect
swimmers and fish consumers; they are designed to protect drinking
water consumers and fish consumers. EPA publishes separate
recreational water quality criteria to protect swimmers. The human
health water quality criteria methodology is based on the source water
protection principle that the polluter should pay for pollution control
rather than the downstream drinking water customer. EPA should have

• TSCA Section 6(b)(4)(D) requires
EPA, in developing the scope of a
risk evaluation, to identify the
hazards, exposures, conditions of use,
and potentially exposed or
susceptible subpopulations the
Agency "expects to consider" in a
risk evaluation. This language
suggests that EPA is not required to
consider all conditions of use,

Page 161 of 212

-------


evaluated the impact of 1,4-dioxane wastewater discharges on the
quality of source water for public water supply systems and been
prepared to find an unreasonable risk if predicted concentrations
exceeded EPA's recommended lifetime drinking water health advisory.
EPA's evaluation of the impact of these discharges on swimming and
fish consumption are appropriate analyses but do not substitute for
analysis of the much higher risk pathway of drinking water.

• Focusing on occasional recreational exposure, the analysis of surface
water fails to satisfy the source water protection goals of water quality
criteria for human health under CWA.

hazards, or exposure pathways in risk
evaluations. EPA has therefore
tailored the scope of the risk
evaluations for 1,4-dioxane using
authorities in TSCA sections 6(b) and
9(b) and focused this fit-for-purpose
evaluation on general population
exposures to ambient water via
recreational activities such as
swimming.

78

• It is well known that 1,4-dioxane is an impurity in a broad range of
personal care and cleaning products used by millions of consumers.
These "down the drain" products contribute 1,4-dioxane to wastewater
and surface water and, together with other sources of this chemical,
account for the widespread presence of 1,4-dioxane in drinking water
throughout the country. Given that 1,4-dioxane is a likely human
carcinogen, is highly soluble in water, and does not readily biodegrade
in the environment, it is critical that the TSCA risk evaluation of this
chemical focus on the impact of wastewater discharges on drinking
water in the U.S. Regulation of pollutant discharges under the Clean
Water Act is based on the need to protect source water for drinking
water utilities so that the costs of pollution are borne by the polluter, not
by the utility. It is very difficult to remove 1,4-dioxane from source
water, and few utilities in the country employ the expensive, energy-
intensive advanced oxidation or other processes needed to remove or
otherwise treat this chemical. It is imperative that the parties responsible
for 1,4-dioxane releases to the environment posing unacceptable risks to
public health be responsible for eliminating those risks, and it is
imperative that the TSCA risk evaluation ensures this happens.

•	In its evaluation of the ambient water,
general population pathway, EPA
focused its analysis using releases
from the scoped industrial and/or
commercial conditions of use shown
in Table 2-2 of the final risk
evaluation. These were based on
reasonably available 1,40-dioxane
release data. It also incorporated
monitoring data that were submitted
during the public comment period
and SACC review of the draft risk
evaluation.

•	Section 1.4.2 of the risk evaluation
describes exposure pathways and
risks that fall under the jurisdiction of
other EPA-administered statutes. As
described in Section 1.4.2 of the risk
evaluation, EPA believes it is both
reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific

82, 85

• In the Supplemental Analysis, US EPA expressly recognizes that "1,4-
Dioxane exposures to the general population may occur from the
conditions of use due to releases to air, water or land." Nonetheless, it
admits that it "did not evaluate unreasonable risk to the general

Page 162 of 212

-------


population from ambient air, drinking water, and sediment pathways for
any conditions of use in this risk evaluation, and the draft unreasonable
risk determinations do not account for exposures to the general
population from ambient air, drinking water, and sediment pathways."

•	EPA wrongfully asserts that it need not evaluate general population and
other exposures because such exposures might be covered under other
environmental statutes administered by EPA.

•	The supplemental Analysis does not identify any authority that would
allow US EPA to disregard its TSCA obligations merely because they
may overlap with obligations under other environmental laws. Reading
TSCA in this light would have the effect of rewriting the requirement
that US EPA conduct risk evaluations "without consideration of costs or
other nonrisk factors."

environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1).

83

• EPA ignored much of its own data obtained through the Toxics Release
Inventory (TRI) in its ambient water exposure analysis. EPA's
Enforcement and Compliance History Online (ECHO) Water Pollution
Search database provides significant data ignored in EPA's analysis.
EPA arbitrarily chose to limit its analysis to facilities fitting select
Occupational Exposure Scenarios it developed for the risk evaluation
(EPA Table 2-4), leading it to omit many emitting facilities as well as
contaminated watersheds and resulting in an underestimation of both the
extent and magnitude of 1,4-dioxane discharged to ambient water. The
ECHO database demonstrates that in 2018, numerous U.S. watersheds
received discharges of 1,4-dioxane that EPA did not include in its
analysis. These include three watersheds receiving more 1,4-dioxane
than those that EPA used for its analysis of ambient water exposures.
These watersheds and their associated facilities are: Spring Creek-Mud
Creek in Alabama (Indorama Ventures), Kendrick Creek-South Fork
Holston River in Tennessee (Eastman Chemical Co Tennessee
operations), and Back Creek in North Carolina (Starpet Inc.). APG
Polytech LLC also had a greater discharge, although its associated
watershed is not reported in the public database. The watershed
receiving the greatest discharges in 2018, Spring Creed-Mud Creek in

•	The modeled releases are based on
the evaluated occupational exposure
scenarios {i.e., industrial and/or
commercial conditions of use, see
Table 2-2 in the final risk evaluation)
and are not intended to reflect or
capture contributions from other
industrial, commercial, or consumer
sources.

•	EPA thanks the commenter for
flagging this potential transcription
error. However, EPA utilized the
entirety of the mass discharged to
publicly owned treatment works
(POTWs), which sums to 37,304
lbs/yr, for the purposes of assessing
releases for the referenced site (Suez
WTS Solutions USA Inc.).

Page 163 of 212

-------


Alabama, had indirect discharges more than six times higher than the
watershed EPA relied on as its upper bound (Ninemile Creek polluted
by Suez WTS Solutions USA Inc.) based on the 2018 data. EPA also
ignored 1,4-dioxane discharges into three other watersheds: Back River-
Cooper River in South Carolina (Dak Americas LLC Cooper River
Plant), Singleton Swamp in South Carolina (Nan Ya Plastics Corp
America), and Brushy Creek-Enoree River in South Carolina
(Mitsubishi Polyester Film Inc.).

• There appears to be a major transcription error in Table 2-4 for the
releases into Ninemile Creek from Suez WTS Solutions USA Inc.

EPA's error led it to use a higher discharge than actually reported in the
2018 TRI (see footnote 44). However, this should not reduce the upper
bound of EPA's predicted surface water concentrations; rather, EPA
should use the significantly higher level discharged by Indorama
Ventures into Spring Creek-Mud Creek, Alabama.



83

•	EPA has not provided sufficient analysis or explanation to support its
selection of concentrations to represent 1,4-dioxane in ambient water.
EPA included monitoring data from Minnesota (MN) and North
Carolina (NC) state agencies because the data were submitted during the
2019 comment period. Although data from NC, specifically Haw River
and Cape Fear River, and MN were the only data submitted during the
comment period, they are not the only states with available data on 1,4-
dioxane concentrations in surface water. While EPA appears to have
used the concentrations at the upper end of the range of the MN and NC
monitoring data, there is a lack of transparency regarding why only
these data were used. The agency stated (on p. 28) that its predicted
values (modeled using E-FAST) for surface water concentrations taken
as a whole cover a range that encompasses the ranges reported by NC
and MN; however, EPA did not explain why it excluded other available
data.

•	EPA inappropriately excluded data points from its STORET database.
Our review of the STORET and NWIS data between the years 2009-

•	In its evaluation of the ambient water,
general population pathway, EPA
focused its analysis using releases
from the scoped industrial and/or
commercial conditions of use shown
in Table 2-2 of the final risk
evaluation. These were based on
reasonably available 1,40-dioxane
release data. It also incorporated
monitoring data that were submitted
during the public comment period
and SACC review of the draft risk
evaluation.

•	EPA thanks the commenter for
pointing out this consideration of the
STORET data cited in the draft risk
evaluation. While EPA did consider
the referenced range from STORET

Page 164 of 212

-------


2019 revealed that there were 59 samples with method detection limits
(MDLs) greater than 100 ug/L. Of these, 34 were from surface water
samples, some of which had MDLs as high as 28,000 ug/L-nearly six
times higher than the upper end of the range of values in the Supplement
that EPA derived using E-FAST (p. 28). EPA discarded all of these
values -even though the "true" concentration of these surface water
samples may be well above its E-FAST derived values EPA uses to
estimate risk. In such cases, researchers would typically either use the
MDL or a value that is one-half of the MDL. EPA simply eliminated the
data from consideration without basis.

• 1,4-Dioxane was measured in various water systems from 2013-2015 as
part of the Third Unregulated Contaminant Monitoring Rule (UCMR3).
The UCMR3 provide relevant 1,4-dioxane water concentration data
from multiple states (US EPA, 2016).

for years 2007 through 2017 in its
draft risk evaluation, the RQs
presented in Table 5-2 are dependent
on results of the screening-level
modeling analysis. Additionally,
though some of the sampling MDLs
were higher than the chronic COC,
EPA did not use unreported/unknown
levels below such MDLs as the basis
for an RQ or unreasonable risk
determination. In response, EPA has
augmented its discussion of
uncertainty in Section 4.3.2 with a
discussion of this point.

• Data from UCMR3, which provides
nationally representative data on the
occurrence of contaminants in
drinking water, were not utilized in
the aquatic exposure assessment due
to a focus on ambient surface water
levels since general population
drinking water exposures were not
included in the scope of the risk
evaluation (see Section 1.4.2).

90

• The draft supplemental analysis includes an evaluation of general
population exposures to 1,4-dioxane from recreational activities {i.e.,
swimming) in ambient surface water, using modeled surface water
concentrations based on E-FAST modeling and measured surface water
concentrations. The modeled concentrations are based on incidental
exposure to 1,4-dioxane in surface waters (swimming) downstream of
industrial discharges (Table 2-6). The measured concentrations are from
three reports: Sun et al. (2016), North Carolina DEQ, and Minnesota
DEQ. Based on the concentrations observed, Sun et al. (2016) suggests

• The modeled releases are based on
occupational exposure scenarios {i.e.,
industrial and/or commercial
conditions of use) and are not
intended to reflect contributions from
the use and/or disposal of consumer
products. The modeling captures
reported releases from industrial
facilities or WWTPs that directly

Page 165 of 212

-------


that one or more industrial sources overshadowed the contribution from
1,4-dioxane that can be expected from the use of consumer products.
However, EPA neglected to cite and use Simonich et al. (2013), which
calculates surface-water concentrations based on measured
concentration from municipal WWTP.

release or transfer 1,4-dioxane.
Similarly, the monitoring data that
were submitted during the draft's
public comment period and SACC
review were utilized but relative
contributions from specific industrial
and/or consumer sources of 1,4-
dioxane are unknown.

83

• The uncertainties (in general population and consumer exposure
estimates) arise in large measure from EPA's failure to have used its
TSCA information authorities to require the development and
submission of the information EPA needed to inform its exposure
assessments. This failure has been raised repeatedly to EPA by
stakeholders over the past several years.

• EPA had sufficient information to
complete the 1,4-dioxane risk
evaluation using a weight of
scientific evidence approach. EPA
selected the first 10 chemicals for
risk evaluation based in part on its
assessment that these chemicals
could be assessed without the need
for regulatory information collection
or development. When preparing this
risk evaluation, EPA obtained and
considered reasonably available
information, defined as information
that EPA possesses, or can
reasonably obtain and synthesize for
use in risk evaluations, considering
the deadlines for completing the
evaluation.

livprorimis C omments

90

• ACC recognizes that this draft supplemental analysis was developed in
response to public and peer review comments on the draft risk
evaluation for 1,4-dioxane. However, including trace levels of
byproducts and impurities in TSCA risk evaluations should not be
routine, given that byproducts and impurities are by definition not
intentionally added or present for commercial purposes, and are often

• TSCA defines "condition of use"
under section 3(4) as "the
circumstances, as determined by the
Administrator, under which a
chemical substance is intended,
known, or reasonably foreseen to be

Page 166 of 212

-------
not present at significant levels. The risk evaluation process must
continue to allow EPA to focus its resources on the conditions of use
that present the greatest potential for risk.

manufactured, processed, distributed
in commerce, used, or disposed of."
Further, EPA explained in the Risk
Evaluation Rule that: "In exercising
its discretion under section
6(b)(4)(D), EPA believes it is
important for the Agency to have the
discretion to make reasonable,
technically sound scoping decisions
in light of the overall objective of
determining whether chemical
substances in commerce present an
unreasonable risk. For example, EPA
intends to exercise discretion in
addressing circumstances where the
chemical substance subject to
scoping is unintentionally present as
an impurity in another chemical
substance that is not the subject of the
pertinent scoping. In some instances,
it may be most appropriate from a
technical and policy perspective to
evaluate the potential risks arising
from a chemical impurity within the
scope of the risk evaluations for the
impurity itself. In other cases, it may
be more appropriate to evaluate such
risks within the scope of the risk
evaluation for the separate chemical
substances that bear the impurity.
(EPA has previously taken an
analogous approach, in requiring
chemical testing of certain chemical

Page 167 of 212

-------




substances under 40 CFR part 766,
based on the potential for the
chemical substance to be
manufactured in such a manner as to
be contaminated with dioxins). In
still other cases, EPA may choose not
to include a particular impurity
within the Scope of any risk
evaluation, where EPA has a basis to
foresee that the risk from the
presence of the impurity would be
'de minimis' or otherwise
insignificant." 82 FR at 33730.

Consumer Kxposurc C omments

89,
83,
86

•	Findings reported by IRIS and ATSDR demonstrate the prevalence of
background concentrations of 1,4-dioxane in the indoor environment -
in some cases at levels that exceed EPA's cancer risk benchmark. As
with other volatile substances like TCE and PCE, 1,4-dioxane's
widespread presence in indoor air is evidence of chronic exposure by
consumers and adds to the dermal and inhalation exposures resulting
from consumer product use.

•	EPA fails to account for background exposure from personal care and
cosmetic products despite SACC reporting that "[t]he decision not to
further analyze background levels of 1,4-Dioxane in any matrix . . .
cannot be supported by any risk assessment principle. Any current use
scenarios increase exposures over those currently being experienced."

•	If EPA ignores the contribution of this background exposure and bases
its risk determinations solely on exposure to TSCA-regulated products,
it will underestimate the true cancer risk to consumers. In quantifying
background exposure, EPA must factor in consumption levels for
different subpopulations, particularly adult women, who are most highly
exposed to 1,4-dioxane from personal care products and cosmetics.
EPA must also factor in increased exposures by low wage workers and

•	The approach of considering
consumer exposures on a product-
specific basis without the
consideration of background or
multiple sources of exposure is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized. This aspect of the
evaluation is described in the
uncertainties Section 4.3.2.

•	Regarding body care and cosmetic
products, they are excluded from the
definition of "chemical substance"
per TSCA section 3(2) and are
outside the scope of this risk
evaluation.

•	As described in Section 1.4.2 of the
risk evaluation, EPA believes it is

Page 168 of 212

-------
communities of color, who are most highly exposed through the use of
cleaning other consumer products and the disposal of wastes containing
1,4-dioxane in or near their neighborhoods.

EPA offers no rationale for its decision to ignore background and multi-
source exposures.

Potentially exposed or susceptible subpopulations (PESS) have not been
properly evaluated - disadvantaged communities can experience higher
levels of dioxane in drinking water, higher incidences of legacy
contamination, and higher concentrations in the products they use.
EPA explicitly calls out that it is ignoring relevant exposures
"particularly for populations living near a facility emitting 1,4-dioxane."
Yet TSCA requires EPA to evaluate a chemical across all of its
conditions of use.

EPA must identify groups near sources of release of 1,4-dioxane as a
potentially exposed subpopulation, given exposures to 1,4-dioxane at
levels higher than the general population.

both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1).

EPA did not consider aggregate or
background exposure that workers,
ONUs, consumers, or bystanders
might be exposed to in addition to
exposures from the conditions of use
in the scope of the risk evaluation
because there is insufficient
information reasonably available as
to the likelihood of this scenario or
the relative distribution of exposures
from each pathway. This may result
in an underestimation of risk, and
EPA acknowledges that risk is likely
to be elevated for individuals who
experience 1,4-dioxane exposure in
multiple contexts. Additional
discussion of this issue has been
added to Sections 4.3.2 and 4.5.
EPA has added a footnote to the
Executive Summary to clarify that
EPA did not identify any legacy uses

Page 169 of 212

-------




of 1,4-dioxane. EPA did not evaluate
"legacy disposal" {i.e., disposals that
have already occurred) in the risk
evaluation, because legacy disposal is
not a "condition of use" under Safer
Chemicals, 943 F.3d 397.

89,

83,

84,
90

•	This document contains limited information on the unintentional
presence of 1,4 dioxane in consumer products. Reported levels of 1,4-
dioxane in consumer products are higher than the supplement assumes.
The concentrations incorporated in the supplement span a broad range
of values and are derived from an extremely limited number of samples,
creating a high degree of uncertainty.

•	EPA provides little information with which to evaluate the few reported
studies on which it relies and does not explain why it failed to include
studies reporting higher levels that were cited in its 2015 problem
formulation and the 2012 ATSDR ToxProfile. EPA did not consider all
publicly available concentration data. The maximum levels reported for
household detergents (160 ppm) are well above the maximum levels the
supplement provides for dishwashing detergent (9.7 ppm) and laundry
detergent (14 ppm). The higher levels would result in risks over 10 X
greater than those calculated by EPA. Recent testing for New York
DEC also reports higher levels for laundry detergent (21 ppm) and
cleaning products (23.1 ppm), again resulting in higher estimates of
exposure and risk.

•	The sources of the 1,4-dioxane concentrations in consumer products
presented by EPA (see Table 2-11) are unclear. The Supplement
indicates that the concentrations of 1,4-dioxane in these consumer
products came primarily from the 2015 TSCA Workplan Assessment.
But the 2015 document does not clearly present the sources EPA used to
derive the concentrations, nor has EPA presented concentration data it
derived from any additional sources it identified through its systematic
review. While the Consumer References Data Screening file includes a
list of articles that EPA presumably used to extract 1,4-dioxane

•	Uncertainties associated with
identifying concentration data for
1,4-dioxane present as a byproduct
are acknowledged and described in
Section 4.3.2. In the acute consumer
exposure scenarios, EPA utilized the
maximum identified weight fraction
in its estimations.

•	EPA conducted a supplemental
literature search per its TSCA
systematic review approach in order
to identify useful information on
consumer exposures, consumer
products, emission rates from
consumer products, etc. The process
is described and shown in Section
1.5.1 of the final risk evaluation.

•	The conditions of use evaluated were
primarily based on those identified in
the 2015 TSCA Workplan
Assessment; however, sourcing of the
product concentration data is shown
in the Supplemental File [Consumer
Exposure Assessment Modeling Input
Parameters]. As described in Section
1.5.1 of the final risk evaluation, a
supplemental systematic review

Page 170 of 212

-------
concentrations in consumer products, the file does not present any
concentration data. Furthermore, EPA does not describe the process it
employed for choosing which concentrations to use in its analysis.
We would like clarification on the range of 1,4-dioxane concentrations
in dish soap (0.7 - 204 ppm). The upper end of this range seems to be
outside of the range with current data for this product category and it
appears to be from a study that is no longer available and is not shown
in supplemental file (Data Quality Evaluation for Data Sources on
Consumer Exposure).

EPA used central tendency weight fractions for chronic exposure
scenarios for high- and moderate-intensity users, rather than the highest
measured concentration for the product class. Because of this approach,
EPA's exposure and risk estimates would not capture products with the
largest levels of 1,4-dioxane and thus would fail to estimate the use
scenarios resulting in the highest risks to consumers, ignoring a
potentially exposed or susceptible subpopulation (PESS) that EPA is
required to consider.

For the supplemental literature search, EPA provides no evidence that it
contacted manufacturers to identify household products containing 1,4-
dioxane or analyzed products for its presence. EPA should require
manufacturers to submit analyses of 1,4-dioxane levels in their products
and, if such data are limited, direct them to conduct product testing
under section 4 of TSCA. At a minimum EPA should compensate for
the inadequate information in its possession by making conservative,
high-end assumptions about the size of the relevant product universe
and the levels of 1,4-dioxane present in these products.

process was carried out in support of
this assessment.

The upper end of the referenced
range (204 ppm) for dish soap was
obtained from the source referenced
in the full comment submission. This
report is accessible through HERO at
https://web.archive.org/web/2009032
0014254/http://www.organicconsume
rs.org/bodycare/DioxaneR.esutts09.pd

f

EPA used central tendency inputs,
including weight fraction, when
estimating lifetime exposures. As
noted in Section 4.3.2, The models
employed (CEM 2.1 and CEM)
typically utilize central tendency
inputs for weight fraction, duration,
frequency, and mass when estimating
lifetime exposures (

1007).

EPA obtained relevant data
consistent with the approaches taken
and peer reviewed in the other
recently finalized solvent risk
evaluations. EPA utilized the
maximum identified weight fraction
in its estimation of acute exposures
and considered and utilized a variety
of sources including peer-reviewed
journal articles, national assessments,
public submissions, and NGO
reports.	

Page 171 of 212

-------
84

The cleaning products industry is trending towards lower levels of 1,4-
dioxane in ethoxylated ingredients and products that contain them. In
California, concentrations of 1,4-dioxane in consumer and commercial
products must be disclosed if greater than 10 ppm per the requirements
of the Cleaning Product Right to Know Act of 2017. As of this writing,
we are unaware of any product that discloses levels of 1,4-dioxane
above the 10-ppm threshold; therefore, all of our member's products are
reasonably assumed to be under 10 ppm, including concentrated
products. Separately, New York has passed a law in 2019 banning the
presence of 1,4-dioxane in cleaning products above 2 ppm in 2022 and 1
ppm in 2023. Manufacturers and suppliers have already initiated efforts
to minimize 1,4-dioxane either by reformulation or via further reduction
in raw materials.

ACI and member companies are working to demonstrate the minimum
method principles that should be in place to detect 1,4-dioxane in
cleaning products at levels appreciably lower than regulatory limits. To
date, we have verified the analytical method principles on five cleaning
product formulations representing hand dishwashing detergents and
laundry detergents. The proposed test formulations (submitted into
docket ad Attachment 1) are meant to be representative of the base
technology used in many products currently on the consumer and I&I
markets, without referencing any one single product or brand. The blend
of materials in the test formulations was chosen to be a benchmark and
represent typical levels of products on the market. The averaged
measured 1,4-dioxane in the products was between 0.94 and 3.6 ppm
(submitted to docket as Attachment 2). Anecdotal measurements from a
company that undertook a cursory landscape scan of cleaning products
found on store shelves found that all tested laundry products had <10
ppm 1,4-dioxane (-70% being < 7 ppm) and top-selling hand dish
products had levels of 1,4-dioxane < 4 ppm.

A recent (November 18, 2020) New York State Department of
Environmental Conservation (NYS DEC) webinar on 1,4-dioxane limits
for household cleaning, personal care, and cosmetic products presented

EPA thanks ACI and HCPA for this
additional context and information on
levels of 1,4-dioxane in consumer
products. The weight fractions
utilized in the evaluation were higher
than those supplied in the
attachments submitted; however, no
unreasonable risks were identified at
the levels modeled. EPA augmented
the uncertainties discussion in
Section 4.3.2 to acknowledge
uncertainty in the weight fractions
applied in the modeling of the dish
soap and laundry detergent consumer
scenarios.

Page 172 of 212

-------


data on a number of cleaning products with a median detection range
from non-detectable to 2.5 ppm.



84,
88,
86,
89

•	EPA chose to only focus on consumer cleaning products without
providing any explanation as to why industrial and institutional cleaning
products were not included. We recommend that EPA evaluate
corresponding industrial and institutional (I&I) and commercial product
conditions of use that were considered for consumers.

•	ACI and HCPA would like further clarification and possible expansion
of the "surface cleaner" scenario. The scenario includes inputs for a
bathroom surface cleaner but does not address other surface cleaners
such as all-purpose cleaners that may be used with some frequency on
multiple surfaces.

•	EPA assumes that the use of surface cleaning products only involves
the use of the inside of one hand, an assumption that understates risk to
any consumer using a surface cleaning product with both hands.

•	The surface cleaner scenario uses a
bathroom as the selected room of use
as a measure of conservatism for
inhalation modeling. The bathroom
has a smaller room volume than other
options such as the kitchen or utility
room. The scenario was not
developed to reflect only bathroom
cleaners - the weight fractions used
reflect the range of identified
concentrations for surface cleaners
(not just bathroom cleaners). A
clarification to this effect was added
to Section 2.4.3.2.1.

•	The hand surface area chosen for the
surface cleaning scenario is
consistent with the surface area used
in the evaluation of other
cleaning/wiping consumer scenarios
for other solvent chemicals recently
evaluated and finalized.

83, 89,
86

• EPA inappropriately dismissed chronic inhalation and dermal exposure
to consumer users from spray polyurethane foam, antifreeze, textile dye,
and paint and floor lacquer. Products such as antifreeze, paints, and
floor lacquer, may be used by "do-it-yourselfers" with more regularity.
Textile dye and other arts-and-crafts products are used regularly by
home hobbyists and artists. While chronic exposure may not be typical
for most consumers, EPA failed to assess DIY users as a "potentially
exposed or susceptible subpopulation." Furthermore, EPA has not
considered exposures arising from gradual release of 1,4-dioxane
following use of such products, for example, from surfaces to which

• The rationale for not estimating
chronic, lifetime exposures to the
referenced products per their
expected intermittent use by
household users is consistent with the
approach taken and peer reviewed for
the other solvent chemicals recently
evaluated and finalized. The
consumer exposure assessment did
not evaluate off-gassing from stored

Page 173 of 212

-------
products as storage of a product
cannot be linked to a condition of use
evaluated and is not in and of itself
identified as a consumer condition of
use within the scope of the Risk
Evaluation. Additionally, TSCA
Section 6(b)(4)(F)(iv) instructs EPA
to factor into TSCA risk evaluations
"the likely duration, intensity,
frequency, and number of exposures
under the conditions of use." This
suggests that activities for which
duration, intensity, frequency, and
number of exposures cannot be
accurately predicted or calculated
based on reasonably available
information were not intended to be
the focus of TSCA Risk Evaluation.

•	The conditions of use evaluated were
primarily based on those identified in
the 2015 TSCA Workplan
Assessment; however, sourcing of the
product concentration data is shown
in the Supplemental File [Consumer
Exposure Assessment Modeling Input
Parameters]. As described in Section
1.5.1 of the final risk evaluation, a
supplemental systematic review
process was carried out in support of
this assessment.

•	The approach of considering
consumer exposures on a product-

	specific basis without the	

Page 174 of 212

they have been applied or during subsequent storage.

•	Adhesives are also not addressed in the supplement. Products with
consumer applications included automotive refinishing coatings, paints,
caulks, sealants, and adhesives. Although they fall within the general
category of paints and coatings identified by EPA, the supplement does
not discuss these specific products.

•	Users of these products likely overlap with users of four household
cleaning products addressed by EPA and thus all products would
contribute to chronic exposure by these consumers, resulting in a greater
cumulative cancer risk.

•	EPA should include consumers who use multiple cosmetics and
cleaning products as potentially exposed or susceptible subpopulations,
given EPA's prior identification of those subpopulations as particularly
likely to be exposed.

-------




consideration of background or
multiple sources of exposure is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized. This aspect of the
evaluation is described in the
uncertainties Section 4.3.2.

• Regarding body care and cosmetic
products, they are excluded from the
definition of "chemical substance"
per TSCA section 3(2) and are
outside the scope of this risk
evaluation.

88, 82,

83,

90

•	EPA does not evaluate potential exposure risk to children under the age
of 11 for acute exposure to 1,4-dioxane in consumer products via
dermal pathways, despite the fact that children may come into contact
with surfaces, dishes and clothing cleaned using products containing
1,4-dioxane, in addition to having access to the cleaning products
themselves. EPA must expand its consumer analyses to include all age
groups, including infants and children, to account for risks associated
with acute or incidental exposures to household consumer products
containing 1,4-dioxane.

•	In finalizing the Draft Risk Evaluation, US EPA should evaluate
products marketed for infants and children to the extent those products
fall under the jurisdiction of TSCA. It should also evaluate exposure
scenarios specific to new parents and young children.

•	Children may be disproportionately exposed to 1,4-dioxane by virtue of
relevant behavioral differences. Relative to adults, children particularly
younger children, engage in significant hand-to-mouth activity and
spend more time on the floor. As a result, children may be
disproportionately exposed to residual 1,4-dioxane present, for example,
from the use of surface and floor cleaners.

• Considering 1,4-dioxane's fate
properties and concentrations of 1,4-
dioxane present in such products
(e.g., laundry detergent and
dishwasher detergent), EPA expects
that any 1,4-dioxane present during
washing will stay with the water and
rinse down the drain rather than
appreciably adhering to surfaces such
as clothing or dishes in any
appreciable amount. The focus on
direct consumer exposures from the
use of products is consistent with the
approach taken and peer reviewed for
the other solvent chemicals recently
evaluated and finalized. Furthermore,
the levels of 1,4-dioxane in the
products is relatively low (up to
0.0009%).

Page 175 of 212

-------


•	EPA needs to identify children as a potentially exposed or susceptible
subpopulation.

•	Evaluation of exposure to the 11 to 15 year-old receptor group from
high-intensity use of antifreeze seems inappropriate.

•	EPA's review of 1,4-dioxane
concentration sources indicated that
the products falling under the
jurisdiction of TSCA are not those
marketed for infants and children.
Such products (e.g., baby shampoos,
lotions) are not under TSCA's
purview and were not evaluated
because the TSCA section 3(2)
definition of chemical substance
excludes cosmetics.

•	The estimation of children's dermal
exposures during the consumer use of
products is consistent with the
approach taken and peer reviewed for
the other solvent chemicals recently
evaluated and finalized. Bystanders
include men, women, and children of
any age.

•	Included children ages 11 and up in
the direct dermal contact scenarios is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized.

88, 83

•	Bystanders, including children, may be present in the rooms where
products containing 1,4-dioxane are used, they may also have dermal
exposure to that chemical through their contact with the surfaces and
dishes that those products are used on or clothes recently washed with a
detergent contaminated with 1,4-dioxane as a byproduct. EPA must
evaluate those known, intended and reasonably foreseen bystander
exposures in its Supplemental Analysis.

•	EPA underestimated consumer bystander exposure by ignoring both

• Generally, individuals that have
contact with liquid 1,4-dioxane
would be users and not bystanders.
Therefore, direct dermal exposures
are not expected for bystanders and
are only estimated for users.
Bystanders include men, women, and
children of any age. EPA's approach

Page 176 of 212

-------


chronic inhalation exposures of this population as well as any dermal
exposures. Bystanders, including young children, will be near sources
(e.g., recently cleaned surfaces, recently washed clothes) on a continual
basis. While their exposure may be lower than the user, lower exposure
is not equivalent to zero exposure, as EPA has effectively assumed. By
ignoring the potential for chronic inhalation exposures to bystanders,
EPA has underestimated risk to this population.

to evaluating bystander exposures is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized.

88

•	EPA relies on outdated model to estimate home size. Home volume of
492 m3 is outdated and EPA should have used the updated value of 446
m3. Moreover, use of average building size leaves lower-income
populations who often live in smaller homes at risk. The recommended
low-end of housing volume is 154 m3.

•	EPA relies on a Consumer Exposure Model that fails to account for the
increases in breathing rates during and immediately following
pregnancy. Instead, EPA's model relies on central tendency inhalation
rate estimates for individuals across all age groups at light intensity
activity levels. EPA's failure to consider for pregnancy- and
postpartum-specific inhalation rates could significantly underestimate
acute inhalation exposure to consumer products containing 1,4-dioxane
in pregnant women and the developing fetus. To better evaluate
exposure risks during vulnerable periods of development, EPA must
consider pregnant women and the developing fetus as potentially
exposed or susceptible subpopulations in this supplemental analysis.

•	The home volume applied in the
modeling of inhalation exposures is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized. However, EPA will
consider the noted update in future
evaluations.

•	EPA's models can account for age-
specific breathing rate differences
when estimating doses. However, in
this evaluation, EPA utilized the 8-
hour maximum time weighted
average air concentrations for users
and bystanders consistent with the
approach taken and peer reviewed for
the other solvent chemicals recently
evaluated and finalized.

79

• The Draft Supplemental Analysis to the Draft Risk Evaluation for 1,4-
Dioxane reviews potential consumer exposure to 1,4-Dioxane in SPF
products. SFC agrees that the characterization of 1,4-Dioxane as a
byproduct is correct. In the Draft Supplemental Analysis, EPA states
that concentrations of 1,4-Dioxane in SPF formulations range between
<0.5 to 500 ppm. While SFC's members that manufacture SPF believe
EPA's estimated concentrations are high, due to the short comment
period, SFC could not verify the exact potential concentrations.

• While application of SPF insulation
products may primarily be
occupational, a "do it yourself' or
DIY installation of SPF is possible.
There are consumer products
available that may expose consumers
(users and bystanders) to 1,4-dioxane

Page 177 of 212

-------


•	Moreover, EPA does not clearly define what types of SPF products are
available for consumer application and, therefore, are included in the
condition of use. There are 4 major types of SPF products. The products
are generally distinguished by their packaging and application
equipment. Figure 1 provides an overview of the four types of SPF
products, their packaging, and their application equipment. High-
pressure SPD are not available for consumer use, unlike low-pressure
and one-component SPF products. Each SPF product has unique
application equipment and exposure profiles. The types of SPF products
available for consumer application are all bead applied, not sprayed. The
supplemental analysis developed emission rates based on high-pressure
SPF. The exposure profiles for one-component and low-pressure SPF
are significantly lower than high-pressure SPF and the final risk
evaluation should consider this distinction. SFC is not aware of any
additional data to inform potential exposure to 1,4-dioxane from these
consumer-type SPF products.

•	PPE and product safety recommendations protect consumer and
occupational applicators. EPA should consider PPE and safety
recommendations when evaluating this condition of use.

•	EPA acknowledges in Section 4.3.2
that the emission rate used is derived
from occupational-grade SPF
products and that there is some
uncertainty about the application of
such data to consumer exposures.
Additional language was added to
further acknowledge the points made
here.

•	EPA does not assume that consumer
users or bystanders will use PPE.

This approach for consumer
conditions of use is consistent with
the approach taken and peer reviewed
for the other solvent chemicals
recently evaluated and finalized.

82

•	The Supplemental Analysis states that US EPA "did not identify any
'legacy uses' or 'associated disposal' But there are numerous scenarios
under which consumers could be exposed to legacy uses. The
Supplemental Analysis observes that 1,4-Dioxane is used in latex wall
paints and in spray polyurethane foams. Consumers could be exposed
to such paints and foams long after their initial application.

•	The Supplemental Analysis states that US EPA did not evaluate "legacy
disposal" (i.e. disposals that have already occurred)" but does not
explain whether US EPA identified such legacy disposals. The failure
to address this point leaves open the possibility - particularly in light of
the legacy exposure scenarios described above - that US EPA may have
identified legacy uses but misclassified them as legacy disposals. US
EPA should address this possibility by providing further explanation of
the reasons it believes there would be no legacy uses of 1,4-Dioxane and

• EPA has added a footnote to the
Executive Summary to clarify that
EPA did not identify any legacy uses
of 1,4-dioxane. EPA did not evaluate
"legacy disposal" {i.e., disposals that
have already occurred) in the risk
evaluation, because legacy disposal is
not a "condition of use" under Safer
Chemicals, 943 F.3d 397. EPA did
not specifically identify any legacy
disposals.

Page 178 of 212

-------


by fully discussing which if any legacy disposals it identified but
declined to consider.



83

• EPA's acute inhalation risk estimates (MOEs) for both users and
bystanders of spray polyurethane foam (SPF) in basements -317 and
384, respectively -are very close to its benchmark MOE of 300; see
Table 4-3 on p. 63 and Table 4-8 on p. 68. EPA does not acknowledge
this, however, or address how the sources of uncertainty it identifies
(pp. 49-54) have been reflected in its decision not to regard this close
margin as indicative of unreasonable risk. It should be noted that EPA's
risk estimate assumes a given consumer has no other exposure to 1,4-
dioxane whatsoever: no dermal exposure through SPF use; no exposure
from the use of another consumer product; no exposure at work; no
exposure through ambient water, no exposure through drinking water;
no exposure from background sources. Otherwise what EPA asserts is
not an unreasonable risk could quickly become one. These examples
show just how narrow and arbitrary EPA's approach to addressing this
chemical's risk is, and how hard it has had to work to avoid finding any
unreasonable risk.

•	EPA considers the uncertainties
associated with each condition of use,
and how the uncertainties may result
in a risk estimate that overestimates
or underestimates the risk. Based on
such analysis, EPA determines
whether or not the identified risks are
unreasonable. Such consideration
carries extra importance when the
risk estimates are close to the
benchmarks for acute, chronic non-
cancer risks, and cancer risks.

•	EPA does not believe exposures need
to be integrated for workers with the
estimated general population
exposures, as the exposures estimated
to be experienced by workers are
generally significantly higher than
general population exposures

Occiipiil ioiiiil r.xposuiT Com incuts

85

• EPA discounts the risk to workers on the assumption that workers will
use personal protective equipment ("PPE") and that the PPE will protect
against 1,4-dioxane exposure. EPA states that it "expects there is
compliance with federal and state laws, such as worker protection
standards, unless case-specific facts indicate otherwise, and therefore
existing [Occupational Safety and Health Administration (OSHA)]
regulations for worker protection and hazard communication will result
in use of appropriate PPE consistent with the applicable [safety data
sheets] in a manner adequate to protect workers." However, EPA
provides no evidence that PPE in the workplace is in fact used and
effectively protects against 1,4-dioxane exposure. Indeed, OSHA itself

• OSHA provides an annotated list of
PELs on its website, including
alternate exposure levels. For 1,4-
dioxane, the alternates provided are
the California OSHA PEL of 0.28
ppm and the ACGIH TLV of 20 ppm.
(https://www.osha.gov/dsg/annotated
-pels/tablez-l.html). EPA's approach
for developing exposure assessments
for workers and ONUs is to use the
reasonably available information and

Page 179 of 212

-------


has recognized that many of its 1,4-dioxane standards are "outdated and
inadequate for ensuring the protection of worker health." EPA must
consider whether 1,4-dioxane presents an unreasonable risk to exposed
workers without discounting that risk by assuming the use and
effectiveness of PPE.

• To the extent that EPA corrects its deficiencies in the Supplement to
include workers exposed to products containing 1,4-dioxane as a
byproduct (industrial laundries and dry cleaners, commercial vehicle
washing, motor vehicle repair and maintenance, cleaning services,
construction and painting), EPA must not distort OSHA standards or
assume universal and effective use of personal protective equipment
(PPE). Likewise, EPA should not assume adherence with
recommendations included in safety data sheets (SDSs), which are not
mandatory, often of insufficient quality to be useful and frequently not
understood.

expert judgment. When appropriate,
in the risk evaluation, EPA has used
exposure scenarios both with and
without engineering controls and/or
PPE that may be applicable to
particular worker tasks on a case-
specific basis for a given chemical.
Thus, while EPA has evaluated
worker risk with and without PPE, as
a matter of policy, EPA does not
believe it should assume that workers
are unprotected by PPE where such
PPE might be necessary to meet
federal regulations, unless it has
evidence that workers are
unprotected. For the purposes of
determining whether or not a
condition of use presents
unreasonable risks, EPA incorporates
assumptions regarding PPE use based
on information and judgment
underlying the exposure scenarios.
These assumptions are described in
the unreasonable risk determination
for each condition of use, in section
5.2. Further, in the final risk
evaluation for 1,4-dioxane, EPA has
determined that most of the industrial
and commercial conditions of use
pose an unreasonable risk to workers
even with the assumed PPE.

88, 83

• The Supplemental Analysis considers only consumer uses of the
products, despite the fact that all of them have known, intended, or

• In response to peer review and public
comments, EPA evaluated eight

Page 180 of 212

-------
reasonably foreseen occupational uses as well. Workers use those
products more frequently, for longer durations, and often in greater
amounts than consumers do, and thus face greater exposures and risks
than the average consumer. Those workers are thus a "potentially
exposed or susceptible subpopulation," whose health EPA is required to
take into account.

There are more than 2.3 million janitors and building cleaners in the
United States, in addition to nearly 500,000 hotel housekeepers. These
workers routinely work with cleaning products such as surface cleaners.
However, EPA did not consider the risks to these cleaning workers in
the Supplemental Analysis. Cleaning workers are exposed to, and
harmed by, chemicals in cleaning products. Exposure to 1,4-dioxane has
the potential to exacerbate those respiratory harms and subject workers
to a range of other serious risks, including cancer, liver disease, and
more.

In the Supplemental Analysis, EPA found even consumer use of surface
cleaners containing 1,4-dioxane would result in risks equal to EPA's
unreasonable risk threshold of one additional cancer for every 1,000,000
people. Had EPA considered workers' greater exposure to surface
cleaners, it would have calculated cancer risks far beyond the
unreasonable risk threshold, requiring EPA to regulate 1,4-dioxane "to
the extent necessary" to eliminate those unreasonable risks. EPA's
failure to consider occupational uses of cleaning products violates
TSCA and leaves millions of workers potentially exposed to unsafe
levels of a carcinogen.

There are also more than 2.2 million domestic workers nationwide,
including more 343,000 paid house cleaners. More than ninety-five
percent of those house cleaners are women, and nearly sixty percent are
Latina. Twenty-five percent of house cleaners have household incomes
below the poverty line, compared to five percent of non-domestic
workers, and less than eight percent of house cleaners receive employer-
provided health insurance, compared to forty-eight percent of non-
domestic workers. These socioeconomic stressors render those workers

consumer uses of products that
contain 1,4-dioxane as a byproduct.
EPA made a policy decision in
consideration of 1,4-dioxane as a
byproduct, to limit consideration to
consumers. TSCA Section 6(b)(4)(D)
requires EPA, in developing the
scope of a risk evaluation, to identify
the hazards, exposures, conditions of
use, and potentially exposed or
susceptible subpopulations the
Agency "expects to consider" in a
risk evaluation. This language
suggests that EPA is not required to
consider all conditions of use,
hazards, or exposure pathways in risk
evaluations. EPA has therefore
tailored the scope of the risk
evaluations for 1,4-dioxane using
authorities in TSCA sections 6(b) and
focused this fit-for-purpose
evaluation on consumer (and
bystanders) exposures to household
products containing 1,4-dioxane as a
byproduct.

As described in Section 4.3.2 of the
final risk evaluation, inhalation and
dermal exposures were evaluated on a
product-specific basis and are based
on use of a single product type within
a day, not multiple products. EPA
does not believe exposures need to be
integrated for workers with the

Page 181 of 212

-------
more susceptible to the impacts of chemicals like 1,4-dioxane. Domestic
workers' exposures to cleaning products containing 1,4-dioxane are far
greater than the average consumer's. EPA's Supplemental Analysis
does not attempt to measure those exposures, much less the
corresponding risks. Moreover, domestic workers may work with, and
be exposed to, multiple types of products containing 1,4-dioxane. In a
single shift, a house cleaner or maid may use surface cleaners, dish
soap, dishwasher detergent and laundry detergent. Because EPA
evaluates the risks from all of those products separately, however, it
fails to account for the combined risks to individuals who are exposed to
all of them simultaneously. This omission further understates the risks
to workers and violates EPA's statutory obligations.

1,4-dioxane also has been detected in latex wall paint and floor lacquer.
There are more than 379,000 house painters nationwide, and those
workers "are potentially exposed to the chemicals found in paint
products during their application and removal." In other risk
evaluations, EPA evaluated painters' occupational exposures to
methylene chloride, perchloroethylene, and pigment violet 29. However,
EPA did not perform that analysis for 1,4-dioxane. EPA's exposure
assumptions do not attempt to capture the known, occupational uses of
paints and floor lacquer. EPA assumes that all exposures to those
products are acute, as opposed to chronic, because of their allegedly
"infrequent and intermittent use frequencies." For professional painters,
however, the use of latex wall paint is neither infrequent nor
intermittent; instead, they use paint every day on the job. In EPA's prior
risk evaluations, it considered workers' chronic exposures to the
chemicals contained in paints and coatings. EPA provides no
explanation for its failure to consider those same exposures in the 1,4-
dioxane Supplemental Analysis.

1,4-dioxane is also present in aircraft de-icing fluids and antifreeze. In
its Supplemental Analysis, EPA ignores aircraft de-icing and considers
only consumer uses of antifreeze. EPA did not evaluate chronic
exposure to antifreeze because of the product's allegedly "infrequent

estimated general population
exposures, as the exposures estimated
to be experienced by workers are
generally significantly higher than
general population exposures
The paint and lacquer uses are
evaluated under consumer exposure.
In addition, occupational exposure of
lubricant and 1,4-dioxane mixture in
sprayed applications, and functional
fluids are discussed in the risk
evaluation document and Appendix
G.

EPA initially excluded production of
1,4-dioxane as a byproduct from
certain other chemicals and presence
as a contaminant in industrial,
commercial and consumer products
from the scope of the risk evaluation
using EPA's discretion under TSCA
section 6(b)(4)(D). While EPA has
addressed some conditions of use
related to 1,4-dioxane as a byproduct
in this risk evaluation, EPA expects
that the commercial use of aircraft
de-icing fluid and similar products
would be considered in the scope of a
risk evaluation for ethylene glycol. In
cases like this, EPA believes its
regulatory tools under TSCA section
6(a) are better suited to addressing
any unreasonable risks that might
arise from these activities through

Page 182 of 212

-------
and intermittent use frequencies." But that is not the case for mechanics,
aircraft de-icers, and other workers who routinely use antifreeze and de-
icing fluids. Increased levels of ethylene glycol, the active ingredient in
antifreeze, have been detected in auto mechanics, the result of chronic
usage and exposures. Those same workers and others face chronic
exposure to 1,4-dioxane from antifreeze and de-icing fluids, and EPA
must consider the risks from those exposures in its Supplemental
Analysis.

Workers using surface cleaners, soaps, and detergents, those using
paints, antifreeze, textile dyes, and SPF (e.g., employees of painting
service companies, automotive garages, textile businesses and home
insulation installers) must certainly be considered chronically exposed
to such products. They are clearly exposed more than once a day, for
multiple days per week, and for longer periods of time per exposure
event than EPA assumed for consumers.

regulation of the activities that
generate 1,4-dioxane as an impurity
or cause it to be present as a
contaminant than they are to
addressing them through direct
regulation of 1,4-dioxane. This case-
by-case approach for byproducts
exposures is consistent with the
various scenarios explained in the
Risk Evaluation Rule, 82 FR at
33730.

With respect to commercial and
industrial use of 1,4-dioxane-
contaminated surface cleaners, soaps,
detergents, paints, antifreeze, and
textile dyes, EPA made a policy
decision in consideration of 1,4-
dioxane as a byproduct, to limit
consideration to consumers. TSCA
Section 6(b)(4)(D) requires EPA, in
developing the scope of a risk
evaluation, to identify the hazards,
exposures, conditions of use, and
potentially exposed or susceptible
subpopulations the Agency "expects
to consider" in a risk evaluation. This
language suggests that EPA is not
required to consider all conditions of
use, hazards, or exposure pathways in
risk evaluations. EPA has therefore
tailored the scope of the risk
evaluations for 1,4-dioxane using

Page 183 of 212

-------




authorities in TSCA sections 6(b) and
focused this fit-for-purpose
evaluation on consumer (and
bystanders) exposures to household
products containing 1,4-dioxane as a
byproduct.

88

• The exclusion of drinking water exposures impacts workers as well as
the general public. UAW represents members at a facility in Florida
where 1,4-dioxane was detected in the water used on site. In addition to
drinking water exposures, this water supply is used in eye wash stations
and for various work-related activities, contributing to employee
exposures. Under TSCA, EPA must evaluate all known, intended, and
reasonably foreseen exposures associated with 1,4-dioxane's conditions
of use, including exposures from drinking and process water.

• As described in Section 1.4.2 of the
risk evaluation, EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1). Currently, EPA is
evaluating 1,4 Dioxane through the
SDWA statutory processes for
developing a National Primary
Drinking Water regulation. However,
EPA has not developed CWA section
304(a) recommended water quality
criteria for the protection of aquatic
life or human health for 1,4-dioxane
and therefore evaluated exposures to
aquatic species and the general

Page 184 of 212

-------




population from ambient water in the
1,4-dioxane risk evaluation.

86, 88

•	The expanded scope fails to consider exposure of occupational receptors
(e.g., housecleaners, janitors, dishwashers, commercial launders,
professional painters). The chronic exposure scenarios still assume one
exposure event per day and therefore may not capture users that
continuously use products throughout the day. Occupational exposures
for agricultural workers and workers at wastewater treatment plants that
may inhale 1,4-dioxane are also omitted.

•	EPA estimates that surface cleaners will be used only once per day, for
15-30 minutes. Regardless of whether that assumption is accurate for
consumers (and EPA has not shown that it is), it clearly underestimates
exposures for janitors, hospitality workers, and others who use cleaning
products repeatedly over the course of their work shifts. In the absence
of information to the contrary, EPA must assume at least eight hours of
daily exposures for those workers over the course of a working lifetime,
as it has for other workers who use 1,4-dioxane for commercial and
industrial purposes.

•	TSCA Section 6(b)(4)(D) requires
EPA, in developing the scope of a
risk evaluation, to identify the
hazards, exposures, conditions of use,
and potentially exposed or
susceptible subpopulations the
Agency "expects to consider" in a
risk evaluation. This language
suggests that EPA is not required to
consider all conditions of use,
hazards, or exposure pathways in risk
evaluations. EPA has therefore
tailored the scope of the risk
evaluations for 1,4-dioxane using
authorities in TSCA sections 6(b) and
9(b)(1) and focused this fit-for-
purpose evaluation on consumer
exposures to products containing 1,4-
dioxane as a byproduct.

•	The expanded scope, including use
patterns and other modeling inputs, is
relevant to household consumers and
bystanders and is not intended to
reflect exposures or risks to
occupational or commercial users.

88, 89

• EPA also fails to evaluate the risks to workers who manufacture and
process the foregoing products. Even if 1,4-dioxane is not intentionally
added to those products, it is still present as a byproduct, and workers
who manufacture cleaning products, paints, and other products
containing 1,4-dioxane may still inhale and come into contact with that

• EPA made a policy decision in
consideration of dioxane as a
byproduct, to limit consideration to
consumers. TSCA Section 6(b)(4)(D)
requires EPA, in developing the

Page 185 of 212

-------


chemical. Their risks were not considered in either the draft risk
evaluation or the Supplemental Analysis.

•	Although the supplement partially addresses consumer exposure to 1,4-
dioxane-containing products, it continues to exclude workplace risks
relating to these conditions of use. This is unjustified under TSCA.
EPA cannot address some phases of chemical's life-cycle and ignore
others. Having designated consumer use of these products as TSCA
conditions of use, it cannot fail to evaluate other "circumstances" of
their manufacture, processing and use, including worker exposure that
occurs during these activities.

•	EPA's final risk evaluation must include risks to workers from exposure
that occurs during manufacture, processing and commercial use of
products containing 1,4 dioxane as a byproduct. It is particularly critical
for EPA to examine risks to the large population of workers who use
cleaning products in industrial and commercial facilities, such as stores,
offices, schools, public buildings, warehouses and factories.

scope of a risk evaluation, to identify
the hazards, exposures, conditions of
use, and potentially exposed or
susceptible subpopulations the
Agency "expects to consider" in a
risk evaluation. This language
suggests that EPA is not required to
consider all conditions of use,
hazards, or exposure pathways in risk
evaluations. EPA has therefore
tailored the scope of the risk
evaluations for 1,4-dioxane using
authorities in TSCA sections 6(b) and
9(b)(1) focused this fit-for-purpose
evaluation on consumer exposures to
products containing 1,4-dioxane as a
byproduct.

Drinkills' \Y siler Kxposurc Comments

76

• EPA has the authority under TSCA to control the introduction into the
environment of contaminants such as 1,4-dioxane that degrade water
quality and increase the cost of water treatment. EPA should be
leveraging all the potential regulatory programs available to reduce
exposure and ergo risk across the environmental spectrum.

• As described in Section 1.4.2 of the
risk evaluation, EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)

76, 81

• Exclusion of drinking water risks is inconsistent with the Frank R.

Lautenberg Chemical Safety for the 21st Century Act, which specifically
highlights the potential to contaminate drinking water as a criterion for
prioritization, establishing the need to consider potential drinking water
risks. A reliance on other environmental statutes to address these risks
may result in gaps in protection. The fact that controls may be
implemented under other statutes does not obviate the Agency's
responsibility to limit the introduction or a problematic chemical into
the environment (including public water supplies and groundwater used
by individual household wells).

Page 186 of 212

-------
76, 87

• Exposures from 1,4-dioxane in drinking water should be added to the
risk evaluation because the Agency has not established a Safe Drinking
Water Act (SDWA) regulatory standard (or Maximum Contaminant
Level [MCL]) for 1,4-dioxane. EPA included exposures to the general
population via ambient surface waters in the supplemental analysis
because there is no nationally recommended Ambient Water Quality
Criteria under the Clean Water Act (CWA). The Agency's reasoning to
include ambient water, but not drinking water, is unclear and
inconsistent with the approach provided in the risk evaluation.

and 9(b)(1). Currently, EPA is
evaluating 1,4 Dioxane through the
SDWA statutory processes for
developing a National Primary
Drinking Water regulation. However,
EPA has not developed CWA section
304(a) recommended water quality
criteria for the protection of aquatic
life or human health for 1,4-dioxane.
Human exposure to a receptor using
the waters for recreation
and exposures to aquatic life were
evaluated in this risk evaluation
under TSCA.

• OCSPP has coordinated with the
Office of Water regarding 1,4-
dioxane contamination in drinking
water. In EPA's Preliminary
Regulatory Determinations for
Contaminants on the Fourth Drinking
Water Contaminant Candidate List
(85 FR 14098 (Mar. 10, 2020)), EPA
found that 1,4-dioxane is occuring in
finished drinking water above a
health reference level and therefore,
for purposes of TSCA section 9(b),
EPA has found risk from 1,4-dioxane
contamination at certain levels in
drinking water that could be
addressed under EPA's SDWA

76, 87

• EPA should explain why the Office of Water is relying on the TSCA
risk evaluation to make a regulatory determination for 1,4-dioxane,
when OCSPP is excluding drinking water exposure from its analysis.
ASDWA continues to stress the need to harmonize regulatory
approaches between OCSPP and the Office of Water so that potential
downstream water contamination from chemicals such as 1,4-dioxane is
not left to the state primary agencies and water systems to solve.
Preventing contaminants from entering drinking water sources is more
effective and less expensive than having to remove them from drinking
water has become contaminated. Protecting drinking water sources (and
preventing contamination) is essential for sustaining safe drinking water
supplies, protecting public health and the economy, and protecting the
environment.

81, 87

• The exclusion of drinking water and other risks misplaces the risk
management burden, moving it from manufacturers and commercial
users onto water rate payers, local government, and other regulated
entities. TSCA should support and minimize the need for controls under
other environmental statutes.

89

• By continuing to omit contaminated drinking water from the evaluation,
the supplement fails to account for a source of exposure that contributes
to total consumer intake of 1,4-dioxane and greatly adds to the cancer
risk from use of consumer products. Including drinking water in the
evaluation would require EPA to add ingestion to inhalation and dermal
routes of exposure; the draft risk evaluation and supplement do not

Page 187 of 212

-------


account for this exposure pathway.

authorities.1 However, EPA has
deferred a determination to regulate
1,4-dioxane under SDWA because
SDWA section 1412(b)(l)(B)(ii)
requires that EPA determine after
opportunity for public comment that
regulation of 1,4-dioxane meets all
three criteria for regulation under
SDWA section 1412(b)(1)(A), and
EPA is awaiting new information that
can inform the evaluation of these
three criteria (i.e. adverse effect, level
of public health concern and
meaningful opportunity for health
risk reduction). EPA will continue to
evaluate 1,4-dioxane under SDWA
authorities to determine whether or
not to regulate 1,4-dioxane in
drinking water, and the information
produced in the risk evaluation
process will be considered by the
Office of Water as part of future
SDWA actions.

• As described above, EPA has regular
analytical processes to identify and
evaluate drinking water contaminants
of potential regulatory concern for
public water systems under SDWA.
The Office of Water evaluates the
regulatory determination criteria

89

•	1,4-Dioxane contamination of drinking water is a national concern.
UCMR3 sampling identified 1,4-dioxane levels were above 0.35 mg/L
in at least one sample from 6.9 percent of PWSs, serving a total of 29.4
million customers in 37 states. According to the EPA drinking water
program, 1,4-dioxane levels of 0.35 mg/L (or 1 ppb) represent "the
amount of 1,4-dioxane expected to cause no more than one additional
case of cancer in 1 million people who drink and bathe with the water
over a lifetime."

•	In areas of North Carolina, the maximum concentrations reported - 114
and 107 ug/L - are nearly 300 times greater than EPA's one in a million
cancer risk level, presenting a significant public health concern.

•	The UCMR3-based estimates understate the number of people
consuming 1,4-dioxane in drinking water at levels of health concern
because medium and small water systems may not test regularly for 1,4-
dioxane and private wells are not required to test at all. As with other
volatile compounds like TCE and PCE, water from municipal systems
and private wells is not only ingested but used for bathing and
showering, which result in inhalation and dermal exposure.

89

•	The millions of users of contaminated drinking water in Eastern North
Carolina and similar "hot spots" in other states comprise PESSs because
their exposure is a function of both drinking water consumption and use
of consumer products and therefore is greater than exposure by the
general population.

•	EPA is required under TSCA to make unreasonable risk determinations
for these highly exposed and susceptible subpopulations. Since drinking
water levels in many communities are well in excess of the
concentrations deemed by EPA and state regulators to pose a 1 in 1
million cancer risk, the combined exposure from drinking water and
consumer product pathways in these communities is likely well above
EPA's unreasonable risk benchmark for carcinogenicity.

89, 82

• The supplement continues to assert that TSCA should not apply "when
other EPA offices have expertise and experience to address specific

Page 188 of 212

-------
environmental media" and that excluding drinking water from TSCA
risk evaluations is necessary to "avoid duplicating efforts taken pursuant
to other Agency programs."

However, 1,4-dioxane is NOT being "addressed" under the SDWA.
EPA has not promulgated a National Primary Drinking Water regulation
for l,4dioxane and has no plans to do so and there is little prospect that
1,4-dioxane in ground water or drinking water will be regulated by EPA
for the foreseeable future.

It contradicts the position that US EPA has taken under its Safe
Drinking Water Act ("SDWA") program. Even though US EPA now
states in the Supplemental Analysis that the drinking water impacts of
1,4-Dioxane should be considered under the SDWA rather than TSCA,
it declined, in March of this year, to make a preliminary regulatory
determination under SDWA for 1,4-Dioxane, on the ground it wanted
first to complete its TSCA risk evaluation of 1,4-Dioxane. US EPA
cannot argue, under the SDWA, that it will consider such impacts under
TSCA and then, under TSCA, argue that it will only consider them
under the SDWA.

1,4-Dioxane in ground water is also ignored but is considered a major
source of drinking water contamination in "hot spots" such as Long
Island, New York and Southern California.

If drinking water sources of exposure are not included in the ongoing
TSCA risk evaluation, the risks they present will likely never be
identified, evaluated, and reduced. In short, even if EPA is correct that
TSCA is a "gap-filling" statute, addressing drinking water
contamination in risk evaluations would in fact fill a serious "gap" in
regulatory protection. To ignore this gap would violate the spirit and
letter of TSCA.

under SDWA Section 1412(b)(1)(A)
to determine whether or not to initiate
the development of a National
Primary Drinking Water Regulation.
EPA promulgates National Primary
Drinking Water Regulations
(NPDWRs) under SDWA when the
Agency concludes a contaminant
may have adverse health effects,
occurs or is substantially likely to
occur in public water systems at a
level of concern and that regulation,
in the sole judgement of the
Administrator, presents a meaningful
opportunity for health risk reduction.
For each contaminant with NPDWRs,
EPA sets an enforceable Maximum
Contaminant Level (MCL) as close
as feasible to a health based, non-
enforceable Maximum Contaminant
Level Goals (MCLG) or establishes a
treatment technique. Feasibility refers
to both the ability to treat water to
meet the MCL and the ability to
monitor water quality at the MCL,
SDWA Section 1412(b)(4)(D).

Public water systems are generally
required to monitor for the regulated
chemical based on a standardized
monitoring schedule to ensure
compliance with the maximum

Page 189 of 212

-------




contaminant level (MCL). Under
SDWA, EPA must also review
existing drinking water regulations
every 6 years, and if appropriate,
revise them. SDWA, originally
passed by Congress in 1974, thereby
is the main federal statute to protect
drinking water by regulating the
nation's public drinking water supply
and authorizing EPA to set national
health-based standards and take other
actions to protect against
contaminants that may be found in
drinking water.

• EPA will continue to evaluate 1,4-
dioxane under SDWA authorities to
determine whether or not to regulate
1,4-dioxane in drinking water, and
the information produced in the risk
evaluation process will be considered
by the Office of Water as part of the
current SDWA actions.

89, 81,
83

• EPA's Legal Justification for Including Surface Water Discharges in the
Supplement Conflicts with Its Rationale for Excluding Drinking Water
Contamination. EPA continues to exclude 1,4-dioxane in drinking water
from its evaluation because it is theoretically subject to the SDWA, but
points to the absence of regulation under the CWA to justify addressing
surface water under TSCA.

• EPA is currently evaluating 1,4-
dioxane under SDWA. There is no
such evaluation underway under
CWA and so EPA is including
surface water discharges in this risk
evaluation.

Page 190 of 212

-------


•	AMWA would like further explanation as to why the exclusion of a
standard under CWA calls for the agency to consider exposure via
surface water, but a lack of national standard under SDWA does not.
This seems inconsistent and calls into question the agency's decision to
exclude drinking water.

•	EPA's selective invoking of another statute as a basis for its 1 lth-hour
decision to include the ambient water pathway, while excluding other
relevant pathways, is contradictory and arbitrary and capricious. The
draft TSCA risk evaluation - and now the Supplement - make clear that
the TSCA risk evaluation will ignore drinking water exposures, on the
basis that they are already addressed by the Office of Water under
SDWA, when in fact they are not.

• As described in Section 1.4.2 of the
risk evaluation, EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1). Currently, EPA is
evaluating 1,4 Dioxane through the
SDWA statutory processes for
developing a National Primary
Drinking Water regulation. However,
EPA has not developed CWA section
304(a) recommended water quality
criteria for the protection of aquatic
life or human health for 1,4-dioxane.
Human exposure to a receptor using
the waters for recreation
and exposures to aquatic life were
evaluated in this risk evaluation
under TSCA.

89,

83,

86, 78

• EPA's analysis of surface water discharges is inadequate to achieve the
purposes of water quality criteria under the CWA or to satisfy the
requirements of TSCA. The supplement presents an incomplete picture
of 1,4-dioxane releases to surface water because it overlooks the many
pathways by which it enters water bodies and the resulting

• EPA disagrees, and believes that the
analysis satisfies the requirements of
TSCA. In its evaluation of the
ambient water, general population
pathway, EPA focused its analysis

Page 191 of 212

-------
contamination of drinking water sources. The surface water discharges
analyzed do not include manufacturer of ethoxylated raw materials or
cleaning products, etc. It also ignores down the drain releases following
use of consumer products.

EPA's evaluation of surface water impacts of 1,4-dioxane discharges is
based on only 24 sources, comprising mainly chemical, pharmaceutical
and pesticide manufacturers, that EPA admits are "likely not
representative of all the releases in the U.S. for 2018." The 24 sources
do not include manufacturers of ethoxylated raw materials or finished
cleaning products, personal care products or cosmetics formulated from
these raw materials. In addition, the supplement identifies several
additional groups of 1,4-dioxane dischargers, representing over 1.6
million facilities.

EPA estimates the number of release days per year for these dischargers
and calculates representative discharge levels but makes no attempt to
estimate the resulting surface water concentrations attributable to each
discharger. This is a significant limitation because the large universe of
discharging facilities likely has significant cumulative water quality
impacts that are broadly distributed geographically. Had EPA's analysis
accounted for all of the numerous industrial point-sources of 1,4-
dioxane, its modeling of ambient water levels would necessarily have
reflected the impact of multiple discharges on specific water bodies.

This would be a more realistic scenario than modeling the surface water
impact of individual dischargers standing alone, the approach EPA uses
in the supplement.

The supplement also ignores "down the drain" releases of 1,4-dioxane
following the use of cleaning products, personal care products and
cosmetics. Since 1,4-dioxane is difficult to treat and remove, it often
passes through POTWs to surface waters, where it mixes with point-
source discharges from industrial and commercial sites and
contaminates drinking water sources.

For the significant fraction of households that rely on septic systems for
disposing of residential wastewater, available information indicates very

using releases from the scoped
industrial and/or commercial
conditions of use shown in Table 2-2
of the final risk evaluation. The OES
with non-zero releases included:
manufacturing, industrial uses,
function fluids (open-system), spray
foam application, and disposal. These
were based on reasonably available
1,40-dioxane release data. It also
incorporated monitoring data that
were submitted during the public
comment period and SACC review of
the draft risk evaluation.

Regarding body care and cosmetic
products, they are excluded from the
definition of "chemical substance"
per TSCA section 3(2) and are
outside the scope of this risk
evaluation.

As described in Section 1.4.2 of the
risk evaluation, EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)

Page 192 of 212

-------
limited removal. Evidence shows that the chemical can be readily
transported from septic systems directly into domestic drinking water
wells typically relied on by the same households.

EPA continues to ignore 1,4-dioxane's use in oil and gas production
{i.e., releases of 1,4-dioxane arising from its use or presence in both
hydraulic fracturing fluids and produced water).

EPA's evaluation of general population risks is particularly flawed
because it does not evaluate the chronic drinking water risks of the
surface water concentrations predicted at the point of release. EPA
claims in the supplemental analysis that the agency is not evaluating the
drinking water risks of surface water contaminated with 1,4-dioxane
because the agency is relying on the Safe Drinking Water Act to
regulate drinking water. That claim cannot be justified because EPA has
no drinking water standard for 1,4-dioxane now and has no plans to
develop such a standard in the future.

Public and peer review comments EPA previously received identified
numerous additional omissions related to water exposures that EPA has
now failed to include in its Supplement.

EPA's failed to include all exposures through water - including
drinking water, groundwater, and sediment, as well as ambient surface
water. EPA failed to consider direct exposures to ambient water through
activities such as cooking, bathing or showering.

and 9(b)(1).

Related to specific dischargers, all
specific discharges included in the
summary tables presented in Section
2.2.1.2.3 (Table 2-5) were modeled to
estimate site-specific surface water
concentrations (see Section 2.4.2.1.1
and Table 2-34).

The releases estimated and used to
model predicted surface water
concentrations in support of the
ambient water, general population
exposure pathway were associated
with the scoped occupational
conditions of use (see Table 2-2 for
the OES included in this release
assessment).

With respect to oil and gas
production (hydraulic fracturing),
EPA's review of the FracFocus
reports on 1,4-dioxane indicates that
the 1,4-dioxane is likely present as an
impurity in the ethoxylated alcohols
that are also named in the same
reports. EPA initially excluded
production of 1,4-dioxane as a by-
product from certain other chemicals
and presence as a contaminant in
industrial, commercial and consumer
products from the scope of the risk
evaluation using EPA's discretion
under TSCA section 6(b)(4)(D).
While EPA has addressed some

Page 193 of 212

-------




conditions of use related to 1,4-
dioxane as a byproduct in this risk
evaluation, EPA expects that 1,4-
dioxane exposures associated with
the use of ethoxylated alcohols used
in hydraulic fracturing fluids would
be considered in the scope of a risk
evaluation for ethoxylated alcohols.
In cases like this, EPA believes its
regulatory tools under TSCA section
6(a) are better suited to addressing
any unreasonable risks that might
arise from these activities through
regulation of the activities that
generate 1,4-dioxane as an impurity
or cause it to be present as a
contaminant than they are to
addressing them through direct
regulation of 1,4-dioxane. This case-
by-case approach for byproducts
exposures is consistent with the
various scenarios explained in the
Risk Evaluation Rule, 82 FR at
33730.

89

•	Monitoring studies in North Carolina [a summary of findings are
provided in the comments] demonstrate the widespread impact of 1,4-
dioxane discharges on drinking water quality. The data confirms a
relationship between surface water concentrations and elevated levels of
1,4-dioxane in drinking water, and indicate that most of the measured
levels in surface water were above the EPA and North Carolina
recommended limit of 0.35 ppb for drinking water based on 1,4-
dioxane's carcinogenicity.

•	This analysis would provide a basis for defining the contribution of

• As described in Section 1.4.2 of the
risk evaluation, EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from

Page 194 of 212

-------


manufacturing and processing sites and down-the-drain releases of
consumer products to 1,4-dioxane levels in drinking water and enable
EPA to determine whether these surface water discharges present an
unreasonable risk to the health of drinking water users.

those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1). Currently, EPA is
evaluating 1,4 Dioxane through the
SDWA statutory processes for
developing a National Primary
Drinking Water regulation.

•	EPA has added a footnote to the
Executive Summary to clarify that
EPA did not identify any legacy uses
of 1,4-dioxane. EPA did not evaluate
"legacy disposal" {i.e., disposals that
have already occurred) in the risk
evaluation, because legacy disposal is
not a "condition of use" under Safer
Chemicals, 943 F.3d 397.

•	In March 2020, EPA published a
Preliminary Regulatory
Determinations for Contaminants on
the Fourth Drinking Water
Contaminant Candidate List pursuant
to SDWA authority, see 85 FR
14098. The Agency did not make a
preliminary determination under
SDWA for 1,4-dioxane because the
Agency has not determined whether
there is a meaningful opportunity for
public health risk reduction. EPA will
continue to evaluate 1, 4-dioxane
prior to making a regulatory
determination. The Regulatory

88

• In the Supplemental Analysis, EPA continues to overlook one of the
largest sources of exposure to 1,4-dioxane: contaminated drinking
water. 1,4-dioxane has been detected in thousands of drinking water
supplies serving more than 88 million people, with unsafe levels of the
chemical detected by more than 280 utilities in 26 states. However, EPA
did not consider the risks associated with that drinking water
contamination in either its initial Risk Evaluation or its Supplemental
Analysis. EPA's exclusion of drinking water exposures for 1,4-dioxane
has no legal or factual basis. EPA claims that "exposures to general
population via drinking water ... fall under the jurisdiction of other
environmental statutes administered by EPA," such as the Safe Drinking
Water Act ("SDWA"), and thus need not be considered under TSCA.
But TSCA does not permit EPA to ignore known exposures and risks
resulting from a chemical's conditions of use merely because those
exposures may be regulated under other environmental laws. Moreover
there is no federal limit on 1,4-dioxane levels in drinking water, and just
last March EPA decided against establishing one. EPA cited its ongoing
TSCA risk evaluation to justify its failure to commence a rulemaking
process for 1,4-dioxane under the SDWA, while at the same time
relying on the existence of the SDWA to excuse its failure to consider
drinking water exposures in its TSCA risk evaluation. This regulatory
shell game violates EPA's obligations under TSCA and leaves millions
of people exposed to a known carcinogen in their drinking water.

86

• Drinking water and air can be contaminated from legacy inputs of
dioxane, in addition to ongoing inputs from consumer products. These
were omitted.

86

• Source reduction is recognized as the most effective way to reduce
drinking water contamination. Including drinking water in this
assessment and regulating dioxane's presence in consumer products

Page 195 of 212

-------


would do more to address contamination that the extensive efforts
required to set an MCL.

Determination 4 Support Document
(USEPA, 2019a) and the Occurrence
Data from the Third Unregulated
Contaminant Monitoring Rule
(UCMR 3) (USEPA, 2019b) present
additional information and analyses
supporting the Agency's evaluation of
1,4-dioxane.

• Sections 2.4.2, 4.2.4, and 5.2 for
EPA's evaluation of general
population exposures via the ambient
water pathway.

85

• The draft risk evaluation excludes numerous significant exposure

pathways in which the general population and environment are exposed
to 1,4-dioxane—such as the well-documented risks to those exposed to
contaminated drinking water—thereby understating the overall risk of
1,4-dioxane exposure.

84

• In light of recent concerns raised by stakeholders, including state water
agencies, we recommend that EPA consider evaluating general
population risks associated with drinking water as part of the risk
evaluation.

Aggregate Kxposure C omments

78, 82

•	EPA should aggregate the incidental ingestion and dermal exposure
risks from swimming instead of evaluating them separately.

•	The agency has not aggregated dermal and inhalation exposure to single
products, when that is clearly the situation for consumers. EPA's failure
to combine exposure across these routes results in an understatement of
risk for consumers.

• EPA did not aggregate exposure
across exposure routes (dermal,
inhalation or oral) for occupational,
consumer, or general populations
exposures. EPA chose not to employ
simple additivity of exposure
pathways within a condition of use
because of the uncertainties present
in the current exposure estimation
procedures. There is currently no
PBPK model available to facilitate
evaluation of aggregate exposure
from simultaneous exposure through
inhalation, dermal, and oral contact
with 1,4-dioxane. Without a PBPK
model containing a dermal
compartment to account for
toxicokinetic processes the true
internal dose for any given exposure
cannot be determined, and

Page 196 of 212

-------




aggregating exposures by simply
adding exposures from multiple
routes could inappropriately
overestimate total exposure.
Conversely, not aggregating
exposures in any manner may
potentially underestimate total
exposure for a given individual. EPA
acknowledges in Section 4.3.2 that
the decision not to aggregate risk
could result in an underestimate of
risk.

• This approach is consistent with the
approach taken and peer reviewed for
the other solvent chemicals recently
evaluated and finalized.

89, 82

•	EPA assumes a single use event per day, but many products are used
multiple times (e.g., laundry detergent, dish soap, surface cleaners). Use
has been increased due to the COVID pandemic.

•	EPA needs to recalculate its consumer risk estimates to reflect the
combined effects of concurrent dermal and inhalation exposure, use of
multiple 1,4-dioxane-containing cleaning products simultaneously and
repeated applications of individual products in the course of a day.

•	EPA is able to calculate a cancer risk that it deems "reasonable" only by
artificially considering each 1,4-dioxane source in isolation from others.
However, when all sources are combined to mirror actual real-world
exposure, the cancer risk is clearly much larger than EPA has estimated.

•	Inhalation risks from 1,4-dioxane in spray polyurethane foam used in
basements, attics, and garages (table 4-3) should be considered together,
not room-by-room.

•	EPA concluded that there is
insufficient information to support
analysis of aggregate exposure across
multiple conditions of use. EPA
therefore did not aggregate risk
across multiple consumer products or
uses. EPA assumed a single use event
per day, per the approach taken and
peer reviewed for the other solvent
chemicals recently evaluated and
finalized. EPA acknowledges in
Section 4.3.2 that the decision not to
aggregate risk across conditions of
use could result in an underestimate
of risk.

•	As described in Section 4.3.2 of the
final risk evaluation, inhalation and

Page 197 of 212

-------




dermal exposures were evaluated on
a product-specific basis and are based
on use of a single product type within
a day, not multiple products.

• The analysis of consumer use of SPF
products did not assume application
would occur in all three rooms of use
on the same day; it estimated
exposures for all three home spaces
(attic, basement, garage) to capture
the application scenario likely to lead
to the greatest exposure based on
zone volume and ventilation rate.
These modeled scenarios reflect
distinct exposure activities.

78, 86,
85, 83,
82

•	Consumer exposure receptors are the same as those whose exposures in
ambient water/surface water are assessed following environmental
releases to water. These exposures also should be aggregated with the
COUs. By excluding aggregate exposures from multiple products and
drinking water, this analysis purposefully ignores best available science
and underestimates potential exposures.

•	TSCA requires EPA to "describe whether aggregate or sentinel
exposures to a chemical substance under the conditions of use were
considered, and the basis for that consideration". Here, while EPA has
made clear it hasn't aggregated anything, there is no indication it has
instead applied a sentinel exposure approach - the term never appears in
the Supplement - nor has it described the basis for the approach it has
taken.

•	Concurrent workplace and consumer risks should be aggregated.

•	EPA fails to consider aggregate exposures under the conditions of use
for the general population. The exposures evaluated in the Supplemental
Analysis occur in addition to exposures from air, drinking water, soil,
sediment, and food. For health outcomes with a threshold level of

•	EPA did not aggregate risk across
multiple COUs or pathways. EPA
concluded that there is insufficient
information to support analysis of
aggregate exposure across multiple
conditions of use. EPA acknowledges
in Section 4.3.2 that the decision not
to aggregate risk across conditions of
use could result in an underestimate
of risk.

•	EPA defines sentinel exposure as
"the exposure to a single chemical
substance that represents the
plausible upper bound of exposure
relative to all other exposures within
a broad category of similar or
related exposures (40 CFR Section
702.33)." In this Risk Evaluation,

Page 198 of 212

-------
EPA considered sentinel exposure the
highest exposure given the details of
the conditions of use and the
potential exposure scenarios. Sentinel
exposures for workers are the high-
end scenarios with no assumption of
PPE use within each OES. EPA
considered sentinel exposures in this
Risk Evaluation by considering risks
to populations who may have upper
bound (e.g., high-end, high intensities
of use) exposures. EPA's decision for
unreasonable risk are based on high-
end exposure estimates to capture
individuals with sentinel exposure.

• In accordance with 40 CFR 702.47
".. EPA will determine whether the
chemical substance presents an
unreasonable risk of injury to health
or the environment under each
condition of use within the scope of
the risk evaluation..This approach
in the implementing regulations for
TSCA risk evaluations is consistent
with statutory text in TSCA Section
6(b)(4)(A), which instructs EPA to
conduct risk evaluations to determine
whether a chemical substance
presents an unreasonable risk "under
the conditions of use." As described
in Section 1.4.2 of the risk
evaluation, EPA believes it is both
	reasonable and prudent to tailor	

Page 199 of 212

exposure (commonly assumed for noncancer outcomes), the exposures
evaluated in the Supplemental Analysis could cause consumers to
exceed a threshold even if these exposures considered in isolation do
not. Thus, evaluations that do not consider additivity to other sources of
exposure are incomplete and underestimate health risks by an unknown
margin.

•	EPA is urged to comply with TSCA by considering the risk of impacts
to environmental media and to public health as a result of exposures to
those media.

•	EPA admits that "[background levels of 1,4-dioxane in indoor and
outdoor air are not considered or aggregated in this analysis".

-------




TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1).

• The approach of considering

consumer exposures on a product-
specific basis without the
consideration of background or
multiple sources of exposure is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized. This aspect of the
evaluation is described in the
uncertainties Section 4.3.2

82

• Cancer risks for children 11-15, adolescents 16-20, and adults 21 and
over should be summed and considered as a total lifetime increased
cancer risk (along with calculations of risk from exposure before age
11).

• For inhalation exposures, a lifetime
average daily air concentration
(LADC) was predicted in CEM and
used to estimate cancer risk. For
dermal exposures, a lifetime average
daily dose (LADD) was used to
estimate cancer risk.

89

• Individuals who receive 1,4-dioxane from multiple sources are clearly
PESSs under TSCA by virtue of their elevated exposure.

• EPA agrees that individuals with
multiple sources of exposure may be
potentially exposed or susceptible
subpopulations. EPA has identified

Page 200 of 212

-------




adult and adolescent workers and
ONUs, adult and child consumers
and bystanders, and adults and
children in the general population as
potentially exposed subpopulations.

85

•	EPA's failure to consider 1,4-dioxane exposure from air emissions,
drinking water and groundwater violates TSCA and results in an
assessment of risks to consumers that is incomplete and under-
protective.

•	Subpopulations exposed to 1,4-dioxane from contaminated groundwater
may be exposed to higher levels of 1,4-dioxane than the general
population

• As described in Section 1.4.2 of the
risk evaluation, EPA has determined
that drinking water, air emissions,
onsite releases to land, disposal, and
underground injection pathways fall
under the jurisdiction of other EPA-
administered statutes or regulatory
programs are outside the scope of this
risk evaluation. EPA believes it is
both reasonable and prudent to tailor
TSCA risk evaluations when other
EPA offices have expertise and
experience to address specific
environmental media, rather than
attempt to evaluate and regulate
potential exposures and risks from
those media under TSCA. EPA has
therefore tailored the scope of the
risk evaluation for 1,4-dioxane using
authorities in TSCA Sections 6(b)
and 9(b)(1).

1 In 111:1 n llcnhli llii/iird Comments

78, 89,
83

• EPA's benchmark MOE for chronic effects does not reflect the lack of
data on 1,4-dioxane for critical endpoints. The hazard database for 1,4-
dioxane lacks studies that assess the potential for reproductive and
developmental effects and developmental neurotoxicity (in light of its
known neurotoxic effects in adults). EPA has no dermal toxicity data at
all and for developmental toxicity has only a single short-term study;

• As described above, there is no
universal list of hazard data required
when evaluating chemical risks under
TSCA. Furthermore, for 1,4-dioxane,
EPA has sufficient, reasonably
available hazard information to

Page 201 of 212

-------


hence, the agency lacks any sub-chronic or chronic reproductive,
developmental or neurotoxicity data A 10X UF should be added to
reflect these data gaps, increasing the benchmark MOE to 300.

conduct a risk evaluation and support
the use of the chosen hazard
endpoints. Therefore, EPA did not
use a database uncertainty factor for
hazard in the 1,4-dioxane risk
evaluation.

83, 89

•	The supplement's evaluation of health risks does not include chronic
non-cancer effects. This omission is all the more curious because EPA
identifies such non-cancer effects attributable to 1,4-dioxane exposure
and it calculates and displays corresponding dose-response values for
these effects. Nowhere does EPA provide any rationale or explanation
for this omission.

•	EPA's final evaluation must estimate chronic noncancer risks to
consumers, taking into account all pathways of exposure and
subpopulations with elevated exposure levels.

• EPA evaluated cancer risk for

consumer exposures because it is the
risk driver {i.e., the most sensitive
endpoint) for chronic exposure.
Based on the lack of cancer risk
identified for chronic exposure
through consumer products, EPA did
not further evaluate chronic non-
cancer risks.

83

• With respect to chronic non-cancer dermal effects, in the Supplement
(Table 3-1) EPA indicates it has derived a HED of 1.6 mg/kg/day and
cites two studies for this value: Kociba et al., 1974 and Kasai et al.,
2009. EPA fails provide any explanation of how either or both studies
were used to derive the HED, and it needs to do so. To the extent EPA
relies on the Kasai et al., 2009 study, EPA may need to apply an
additional UFluncertainty factor (LOAEL to NOAEL) as this study did
not identify a NOAEL.

• The supplement was focused on the
addition of new COUs and only
provided a summary final PODs used
to evaluate risk for consumer and
general population exposures.
Complete documentation for
derivation of the final PODs is in
Section 3 and Appendix K of the risk
evaluation. EPA did not apply an
uncertainty factor for NOAEL to
LOAEL extrapolation to the chronic
dermal non-cancer POD because the
POD is based on the BMDL rather
than a LOAEL.

83, 84,
80, 90

• EPA has provided no discussion or description of the basis for its
changes to the POD values and CSF either in the Supplement or in the
other documents it released along with it, other than the brief statement
that the changes were in response to peer review and public comment.

• EPA changed several of the PODs in
response to SACC and public
comments. The supplemental that
was sent for public comment

Page 202 of 212

-------
included a summary table of final
PODs, but, because these changes
weren't the focus of the
supplemental, documentation and
rationale for the changes was not
included. The documentation for the
changes is in the hazard section and
appendices of the final risk
evaluation. EPA's rationale for
changes and responses to previously
submitted comments on these topics
are presented in the preceding
sections of this document. Changes
made to PODs in response to SACC
and public comment include:
The dermal cancer slope factor was
modified based on reanalysis of
female mouse cancer data. This data
was used in the IRIS assessment but
had been initially excluded in the
draft risk evaluation due to difficulty
modeling the data; In response to
comments questioning the exclusion
of this sensitive data, EPA obtained
individual animal data from the
original study to support a more
robust time-to-tumor modeling
approach that allows inclusion of this
data

2. EPA corrected an error in dermal
POD derivation. By incorporating a
dermal adjustment factor in both the
	hazard and exposure portions of the

Page 203 of 212

With one exception, there are no comments that explain the changes.
EPA appears to have made a change consistent with the comments it
received from the New Jersey Department of Environmental Protection
regarding the agency's earlier rejection of the oral cancer slope factor
adopted by IRIS in its 2013 assessment of 1,4-dioxane. EPA has
included this value in the Supplement as the starting point for its
extrapolation to cancer risk from dermal exposure. Aside from this
value, the previous comments from the public or the SACC peer review
report do not explain EPA's decision to change some of the POD
values.

• EPA should make a discussion of the scientific rationale for these
changes available for peer and public review before finalizing the risk
evaluation.

-------




risk calculations of the draft, EPA
had effectively compared PODs in
terms of applied dermal dose to
predicted exposures in terms of
absorbed dermal dose. In the final
RE, EPA has revised all dermal
PODs to reflect absorbed dermal dose
rather than the applied dermal dose
calculated in the draft RE. This
eliminates the error by putting both
the exposure and hazard parts of the
risk equation in terms of absorbed
dermal dose. EPA made this change
in response to SACC comments
indicating an error in the approach.

3. Nasal lesions are now classified as a
result of systemic delivery (as
opposed to portal of entry
effects) relevant for dermal and oral
exposures as well inhalation
exposures. EPA made this change in
response to SACC comments.

80, 90

•	There is considerable scientific evidence that the mode of action (MO A)
of liver and respiratory tract tumors observed in rodents treated with
1,4-dioxane do not arise by a mutagenic mode of action, but instead are
related to a threshold-based response that only occurs at high doses. The
draft supplemental analysis again errs in using a default linear low dose
extrapolation method for calculating theoretical cancer risks. Instead,
EPA should use a threshold approach, or at a minimum, present in
Tables 4.4 and 4.6 risks calculated using a point of departure based on
the threshold MOA side by side with calculations based on unit risk.

•	The decision to use the female mouse liver tumors from the study by
Kano et al. (2009) directly contradicts the 2019 Draft Risk Evaluation

• EPA developed a more thorough
MOA analysis that applies the
framework outlined in the Guidelines
for Carcinogen Risk Assessment to
evaluate evidence for proposed
mechanisms of carcinogenicity for
liver tumors. This analysis is was
substantially revised and expanded
and moved to an Appendix
(Appendix I). The narrative in the
body of the RE is now condensed to

Page 204 of 212

-------
which concluded that "Female mouse hepatocellular carcinoma data
from Kano et al. (2009) were not modeled due to the difficulties that
were previously noted in the U.S. EPA (2013c) IRIS assessment."
o EPA has not explained why it has concluded that these

significant concerns about the female mouse liver tumors are no
longer relevant, nor does the Draft Supplement even indicate
that EPA's view of these tumors has changed,
o The occurrence of female mouse liver tumors is not consistent
with other available information — including the results of a
subchronic study conducted by the same research group (Kano et
al. 2008).

o This conclusion is further supported by the results of a new
subchronic study sponsored by ACC and recently published
(enclosed). A second article describing the results of
toxicogenomic analysis of the livers of the exposed animals has
been submitted for publication and will be available shortly. The
findings of these two analyses lend further support to the
conclusion reached by authoritative bodies around the world that
the tumors observed in the laboratory animal studies are the
result of a threshold mode of action.

Further, the dermal CSF for the female mouse liver tumors in the draft
is roughly the same as the oral CSF for those tumors in the 2013 IRIS
assessment. This discrepancy suggests that EPA has not made the
appropriate adjustment for converting from oral-to-dermal exposure.
The significant difference in the values (two orders of magnitude) for
dermal risks derived from the inhalation study by Kasai et al. between
the 2019 and 2020 drafts also suggests a problem with the approach
taken in the Draft Supplement. It is not clear whether these differences
reflect a change in the approach to extrapolating from oral/inhalation to
dermal exposures or an error in the calculation.

provide a summary of EPA's major
conclusions about MOA. The
fundamental conclusions of this
section have not changed, but the
conclusions are supported by a more
robust analysis. As described in
above responses, EPA considered the
recently published subchronic study
submitted by ACC but did not
incorporate this evidence into the
MOA analysis. While the study may
identify thresholds for specific effects
evaluated in the study, a 90-day study
that does not include tumor endpoints
is not able to demonstrate that the key
events in question are necessary
precursors of liver tumor formation.
The dermal cancer slope factor was
modified based on reanalysis of
female mouse cancer data in Kano et
al (2008). This data was used in the
IRIS assessment but had been
initially excluded in the draft risk
evaluation due to difficulty modeling
the data. In response to comments
questioning the exclusion of this
sensitive data, EPA obtained
individual animal data from the
original study to support a more
robust time-to-tumor modeling
approach that allows inclusion of this
data.

Derivation of the dermal CSF from

Page 205 of 212

-------




the female mouse liver tumors is
presented in Section 3.2.6 and
Appendix K of the risk evaluation.
Because oral absorption was assumed
to be 100%, the dermal CSF is equal
to the oral CSF calculated from the
Kano (2008) data. No additional
adjustment is required for conversion
from an oral dose to an absorbed
dermal dose.

90, 83

•	In table 4.4, EPA indicates that the calculated theoretical cancer risk
value of 1.0 x 10-6 "...exceeds the benchmark of 1 x 10-6." This is
obviously incorrect as these numerical values are equal.

•	EPA fails to identify this unreasonable risk or provide any explanation
for dismissing it. EPA's risk determination for surface cleaners
erroneously states (p. 74): "For consumers, EPA found that there was no
unreasonable risk of non-cancer effects (liver toxicity) from acute
inhalation or dermal exposures or of cancer from chronic inhalation or
dermal exposures at the high intensity use."

•	EPA appreciates this comment. The
language has been slightly revised to
clarify that values may be equal to
the benchmark.

•	EPA's finding for surface cleaners is
based on the acute and chronic
consumer exposure analysis
described in Section 2.4.3.

88, 89,
83

• In both the draft risk evaluation and Supplemental Analysis, EPA

employed a default lOx uncertainty factor to adjust for all human (intra-
species) variability. But EPA has no evidence that differences in
susceptibility to 1,4-dioxane vary only by a factor of 10.
• The Supplemental Analysis falsely asserts that "reasonably available
human health data for all routes of exposure evaluated {i.e., dermal
and inhalation) indicate that there is no evidence of increased
susceptibility for any single group relative to the general
population." This claim is contradicted by EPA's own draft risk
evaluation, which acknowledged that 1,4-dioxane is metabolized at
least in part by Cytochrome P450 ("CYP") enzymes, and that
"[variations in CYP enzyme expression may contribute to
susceptibility." EPA has also acknowledged that these genetic
variations coupled with the effects of certain pre-existing health

• EPA revised the description of PESS
in the executive summary of the final
risk evaluation to be consistent with
more in-depth discussions of PESS
later in the document. In the absence
of quantitative information on the
impact of genetic variability, pre-
existing health conditions, lifestage,
or other factors on susceptibility
across the population, EPA applied
an uncertainty factor of 10 to account
for interindividual variability. In
Section 4.3.6 of the risk evaluation,
EPA recognizes this as a source of

Page 206 of 212

-------


conditions, like fatty liver disease, could alter the metabolism and
toxicological activity of 1,4-dioxane.

•	EPA has previously acknowledged that "a 10-fold factor may ... be
too small because of factors that can influence large differences in
susceptibility, such as genetic polymorphisms."

•	Studies have shown that differences in sensitivity to chemical
exposures can reach up to 30 fold, in some cases up to 100 fold, due
to variability factors like pre-existing health conditions.

•	We recommend that EPA utilize at minimum a 30x intra-species
uncertainty factor for 1,4-dioxane, consistent with the approach
employed by California EPA, where, as here, "differences in
metabolism and excretion are key to the toxicological activity [of a
chemical]," particularly in the context of children's health.

•	We recommend an additional 10X UF for workers and consumers in
recognition of the uncertain range of genetic variability, the very large
worker and consumer populations exposed to 1,4-dioxane, and EPA's
inability to determine the susceptibility to the substance of children and
pregnant women. This would increase the benchmark MOE for non-
cancer chronic health effects to 1000X.

•	While EPA acknowledges "limited data" exist on potential
susceptibilities, it proceeds to ignore them by asserting there is a "lack
of quantitative information." At a minimum, EPA must account for
these susceptibilities in its uncertainty analyses and augment hazard and
risk characterizations to reflect these relevant subpopulations, including
by considering the use of additional uncertainty factors and/or by
adjusting the magnitude of uncertainty factors applied.

uncertainty.

83

•	EPA's reliance on route-to-route extrapolation for sub-chronic/chronic
dermal effects -necessitated by the total absence of dermal toxicity data
-also introduces uncertainty that EPA has failed to account for. As is
recommended for route-to-route extrapolation generally and oral-to-
dermal extrapolation specifically, EPA should apply an additional
uncertainty factor of 10 to account for these uncertainties.

•	EPA also newly examines risks from acute oral and dermal exposure

• EPA agrees that route-to-route
extrapolation is a source of
uncertainty in the hazard
characterization. As described in
Section 3.2.7, EPA concluded that
the primary sources of uncertainty are
likely to underestimate the POD

Page 207 of 212

-------


resulting from contact with 1,4-dioxane contaminated ambient surface
water. Here EPA relies on Mattie et al., 2012, an inhalation study from
which it derived a POD HED of 35.4 mg/kg/day, to evaluate risk. The
associated BMOE of 300 EPA applied does not account for either of the
route-to-route extrapolations employed (inhalation-to-dermal and
inhalation-to-oral). An additional uncertainty factor should apply for the
route-to-route extrapolations employed.

rather than overestimate the POD.
For example, absorption through
lungs is generally expected to be
more efficient for solvents. The oral
and dermal PODs derived under the
assumption of 100% absorption may
therefore be artificially low, but are
unlikely to be artificially high. Given
the cautious assumptions made in the
route-to-route extrapolation, EPA
concluded that an additional
uncertainty factor was not warranted.

83

•	EPA did not examine cancer in assessing risks from acute exposures and
did not provide any explanation for this decision. For dichloromethane
(DCM), the agency explained its decision not to assess acute cancer
risks by stating only that the "[relationship is not known between a
single short-term exposure to DCM [methylene chloride] and the
induction of cancer in humans" (p. 699). We can assume that EPA
would offer the same rationale in the case of 1,4-dioxane. If so, EPA's
rationale is not supported and is unwarranted.

• The National Research Council (NRC) states: "The NRC guidance
states that the determination of short-term exposure levels will
require the translation of risks estimated from continuous long-term
exposures to risks associated with short-term exposures. "

•	EPA did not sufficiently consider such principles related to mode-of-
action in deciding not to model acute cancer risk based on chronic
exposure data. EPA estimates for excess cancer risk were based on the
assumption of linearity in the relationship between 1,4-dioxane
exposure and the probability of cancer. Hence, a linear low-dose
extrapolation from chronic to acute exposures would be the appropriate
approach to take for 1,4-dioxane.

•	A linear extrapolation from chronic cancer bioassays may even
underestimate the cancer risk of short-term exposures. Halmes et al.,

• EPA did not evaluate cancer risks
from acute exposure because the
relationship between acute exposure
and lifetime cancer risk is unknown
and there would therefore be
substantial uncertainty around such
an analysis. This approach is
consistent with the approach taken
and peer reviewed for the other
solvent chemicals recently evaluated
and finalized. EPA applied the
framework outlined in the Guidelines
for Carcinogen Risk Assessment to
evaluate evidence for proposed
mechanisms of carcinogenicity for
liver tumors but did not identify clear
evidence in support of a particular
MOA.

Page 208 of 212

-------


2000 lends support to the potential for short-term exposures to result in
similar or higher cancer risks than even chronic lifetime exposures. The
study used NTP data where both shorter term and full lifetime studies
had been conducted.



I nccrliiinlY Aiiiilvsis Comments

S.i

• JiPA's failure to conduct uncertainty analyses to determine the effects of
its assumptions, limited data, modeling defaults, and so forth. In the
Supplement EPA acknowledges this need. On pages 30 and 49, EPA
states: "EPA's approach recognizes the need to include uncertainty
analysis." But in fact no uncertainty analysis has been conducted. While
EPA identifies factors that contribute to uncertainty, it never evaluates
their effect on its conclusions, and its determinations in effect ignore
these factors. One small exception is that EPA does present a sensitivity
analysis of the consumer exposure model it used, CEM. But that is a
stand-alone analysis of the model itself, and does not extend to a
characterization of the manner in which EPA uses the model outputs to
estimate risks and then make risk determinations for this chemical.

• To the extent possible, EPA describes
the potential magnitude of each
source of uncertainty (see Section
4.3.2). For several sources of
uncertainty, EPA lacks the
quantitative information that would
be necessary to characterize
uncertainty for some parameters.
EPA believes that the qualitative and
quantitative analyses included in the
risk evaluation provide sufficient
information to support risk
conclusions. As discussed in Section
5.1, EPA, in making a risk
determination, takes into account a
number of things, including the
Agency's confidence in the data used
in the risk assessment. This includes
an evaluation of the strengths,
limitations, and uncertainties
associated with the information used
to inform the risk estimate and the
risk characterization.

83

• With regard to consumer exposures, especially chronic, EPA states it
has only moderate confidence in its chronic inhalation exposure
estimates and only low to moderate confidence in its chronic dermal
exposure estimates. Yet it firmly concludes that none of these exposures
presents any unreasonable risk. The basis for these rankings is less than
clear and appears quite subjective. EPA describes the factors it states it
considers in deciding on a confidence ranking, but never shows how
those factors were actually applied to yield the confidence result it
assigns to a specific exposure. At the very least EPA should have
conducted uncertainty analyses to reflect and address uncertainties
engendered by the lack of confidence in the available release and
exposure information.

83

• Overall EPA stated (p. 32): "Based on the above considerations, the
general population ambient water exposure assessment scenarios have
an overall low to moderate confidence." Despite the significant
uncertainties in the available information, however, EPA still draws

Page 209 of 212

-------


unqualified conclusions that human exposures to ambient water (from
swimming and fish consumption) do not present unreasonable risk.
While EPA states that it takes its degree of confidence in available
information into account in making risk determinations (pp. 70, 71),
EPA never explains how it does so; nor does it rationalize or adjust this
particular risk determination with the fact that it has only low to
moderate confidence in the data on which it is based. At the very least
EPA should have conducted uncertainty analyses that reflect and
address uncertainties engendered by the lack of confidence in the
available release and exposure information.



Kdil»ri;il/(

hirilv C omments

82

• In section 4.2.1.2 4.6.2.2 [sic], the last sentence cites sections 2.4.3 and
4.2.3 for methods for consumer exposure assessment and risk
characterization. The Supplemental Analysis, however, does not
include these sections. (The 2019 Draft Risk Evaluation ("Draft Risk
Evaluation") also has no section 2.4.3, and section 4.2.3 covers only
hazard identification.)

• EPA has revised section references to
accurately reflect contents of the final
risk evaluation.

82

•	A footnote to table 3-1 states, "HECs are adjusted from the study
conditions as described above in Section 3.2.6," but there is no section
3.2.6.

•	It seems the new methods sections were not ready and were not
included in the Supplemental Analysis. The existence of such
significant errors calls the overall conclusions of the Supplemental
Analysis into question.

• HEC derivation was beyond the
scope of the Supplemental Analysis
and this footnote should have been
excluded. The final risk evaluation
includes complete documentation of
POD derivation in Section 3.2.6.

82

Several issues of Transparency were identified:

•	There is no discussion of the revisions to the Points of Departure
(PODs) between the Draft Risk Evaluation and the Supplemental
Analysis.

•	The Draft Risk Evaluation describes adjusting PODs for occupational
instead of continuous exposure, but the Supplemental Analysis does not
describe further adjustment of the PODs for consumer exposure.

•	1,4-Dioxane could not be found in the hyperlinked IHSkinPerm©
spreadsheet for the permeability coefficient (5.05E-04 cm/hr). Based on

• POD derivation was beyond the of
the Supplemental document. Some
PODs changed in response to SACC
and public comment. The final risk
evaluation includes complete
documentation of POD derivation in
Section 3.2.6. The rationale for
specific changes made since the draft
in is articulated in above responses to

Page 210 of 212

-------


the link given in table 28, 1,4-Dioxane is not in the pull-down menu of
substances covered by IHSkinPerm©

•	Certain parameter values used to calculate absorption fractions in table
2-12 are not clear. The Consumer Exposure Model 2.1 User Guide
discusses the fraction-absorbed model; however, it does not provide
certain scenario-specific or chemical-specific parameters, such as
temperature or the gas-phase mass transfer coefficient.

•	Lines 1264-1266 cite Frasch and Bunge (2015) for modeling to
estimate the absorption factor based on chemical-specific data. They
modeled four chemicals; none was 1,4-Dioxane. Values used for 1,4-
Dioxane should be provided.

SACC and public comments.

•	IHSkinPerm© allows users to add
new chemicals to the program by
selecting "User's" in the Database
field. Using this option, along with
the physical and chemical properties
for 1,4-dioxane shown in Table 1-1
of the final risk evaluation, a user is
able to estimate the permeability
coefficient referenced.

•	All inputs required to replicate the
runs presented are shown in the
Supplemental File [Consumer
Exposure Assessment Model Input
Parameters], In that file, the gas
phase mass transfer coefficient
estimated and used within CEM 2.1
is shown as 3.2 m/hr.

•	The application of the Frasch and
Bungle (2015) fraction absorbed
estimation to chemicals not specified
in the original source is an approach
that has undergone peer review when
CEM 2.1 was peer reviewed. This
application is consistent with the
dermal modeling conducted and
reviewed in the other recently
finalized solvent risk evaluations.

82

• In table 4-13 of the Draft Risk Evaluation, six dermal cancer slope
factors range from 1.7e-4 to 6.7e-4 per mg/kg-d, depending on the data
and assumptions used. In table 31 of the Supplemental Analysis,
however, the three dermal cancer slope factors range from 1.2e-2 to
1.2e-l, indicating much higher cancer risks than previously estimated.

• POD derivation was beyond the of
the Supplemental document. Some
PODs changed in response to SACC
and public comment. The final risk
evaluation includes complete

Page 211 of 212

-------


This begs the question: which prior determinations of no unreasonable
risk are erroneous? The 2020 Supplemental Analysis does not
acknowledge the underestimate or correct the risk determinations that
appeared in the Draft Evaluation. Without explanation of the revised
PODs, there can be little confidence that the new risk determinations are
not erroneous, too.

documentation of POD derivation in
Section 3.2.6. Revised occupational
risk estimates based on these new
PODs are in Section 4.2. The
rationale for specific changes made
since the draft in is articulated in
above responses to SACC and public
comments.

78

• There is some confusion in the text in Section 2.4.3.3 Consumer
Exposure Modeling Approach as to what model was used for each
scenario. Suggest revising the first sentence to read "Acute exposures
via inhalation and acute and chronic dermal contact to consumer
products were estimated using EPA's Consumer Exposure Model
(CEM) Version 2.1..." "An older version of CEM, available within E-
FAST 2014, was used to estimate chronic inhalation exposures.

• EPA updated this paragraph with this
clarification.

Page 212 of 212

-------