1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
PUBLIC RELEASE DRAFT
December 2024
EPA Document# EPA-740-D-24-026
December 2024
United States Office of Chemical Safety and
Environmental Protection Agency Pollution Prevention
Draft Environmental Hazard Assessment for
Diisobutyl Phthalate (DIBP)
Technical Support Document for the Draft Risk Evaluation
CASRN 84-69-5
ch3
December 2024
-------
PUBLIC RELEASE DRAFT
December 2024
28 TABLE OF CONTENTS
29 ACKNOWLEDGEMENTS 5
30 SUMMARY 6
31 1 INTRODUCTION 7
32 2 APPROACH AND METHODOLOGY 8
33 3 AQUATIC SPECIES HAZARD 9
34 4 TERRESTRIAL SPECIES HAZARD 14
35 5 WEIGHT OF SCIENTIFIC EVIDENCE CONCLUSIONS FOR ENVIRONMENTAL
36 HAZARD ASSESSMENT 17
37 5.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the Environmental
38 Hazard Assessment 17
39 5.1.1 Confidence in the Environmental Hazard Data set 18
40 6 ENVIRONMENTAL HAZARD THRESHOLDS 22
41 6.1 Aquatic Species COCs 22
42 6.2 Terrestrial Species Hazard Values 25
43 7 ENVIRONMENTAL HAZARD ASSESSMENT CONCLUSIONS 25
44 REFERENCES 27
45 Appendix A Analog Selection for Environmental Hazard 32
46 A.l Structural Similarity 33
47 A.2 Physical, Chemical, and Environmental Fate and Transport Similarity 36
48 A.3 Ecotoxicological Similarity 38
49 A.4 Read-Across Weight of the Scientific Evidence and Conclusions 41
50 Appendix B Species Sensitivity Distribution for Acute Aquatic Hazard 42
51 Appendix C Environmental Hazard Details 50
52 C.l Evidence Integration 50
53 C. 1.1 Weight of the Scientific Evidence 50
54 C. 1.2 Data Integration Considerations Applied to Aquatic and Terrestrial Hazard Representing
55 the DIBP Environmental Hazard Database 51
56
57 LIST OF TABLES
58 Table 3-1. Aquatic Organisms Environmental Hazard Studies Used for DIBP, Supplemented with
59 DBP Environmental Hazard Data 11
60 Table 4-1. Terrestrial Organisms Environmental Hazard Studies Used for DIBP 15
61 Table 5-1. DIBP Evidence Table Summarizing the Overall Confidence Derived from Hazard
62 Thresholds 21
63 Table 6-1. Aquatic Environmental Hazard Threshold for DIBP 24
64 Table 6-2. Terrestrial Environmental Hazard Threshold for DIBP 25
65
66 LIST OF APPENDIX TABLES
67 Table_Apx A-l. Structure Program Filtering Criteria 34
Page 2 of 54
-------
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
PUBLIC RELEASE DRAFT
December 2024
TableApx A-2. Structural Similarity between DIBP and Analog Candidates which met Filtering
Criteria in at least 3 out of 4 Structure Programs 35
Table Apx A-3. Analog Candidates with Similar log Kow values to that of DIBP 37
Table Apx A-4. Comparison of DIBP and DBP for Several Physical and Chemical and
Environmental Fate Properties Relevant to Water, Sediment, and Soil 38
Table Apx A-5. Ecotoxicological similarity in aquatic taxa exposed to DIBP (predicted hazard) and
DBP (empirical hazard) 39
Table Apx A-6. Comparison of DIBP and DBP Points of Departure and LC50 Values in Fathead
Minnow Exposed for 24-hours, and LC50 Values in Nitocra spinipes Exposed for 96-
hours 40
Table Apx B-l. Species Sensitivity Distribution (SSD) Model Input for Acute Exposure Toxicity in
Aquatic Vertebrates and Invertebrates - Empirical Data 43
Table Apx B-2. Species Sensitivity Distribution (SSD) Model Input for Acute Exposure Toxicity in
Aquatic Vertebrates and Invertebrates - Web-ICE Data 43
TableApx C-l. Considerations that Inform Evaluations of the Strength of the Evidence within an
Evidence Stream (i.e., Apical Endpoints, Mechanistic, or Field Studies) 53
LIST OF APPENDIX FIGURES
FigureApx A-l. Framework for DIBP Environmental Hazard Analog Selection 33
Figure_Apx B-l. SSD Toolbox Model Fit Parameters 48
Figure Apx B-2. Species Sensitivity Distribution (SSD) for Acute DIBP Toxicity to Aquatic
Vertebrates and Invertebrates 49
Page 3 of 54
-------
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
PUBLIC RELEASE DRAFT
December 2024
ABBREVIATIONS AND ACRONYMS
AF Assessment factor
ChV Chronic value
COC Concentration(s) of concern
EC50 Effect concentration at which 50 percent of test organisms exhibit an effect
EPA Environmental Protection Agency
HC05 Hazard concentration that is protective of 95 percent of the species in the SSD
HV Hazard value
LC50 Lethal concentration at which 50 percent of test organisms die
LD50 Lethal dose at which 50 percent of test organisms die
LOEC Lowest observable effect concentration
LOAEL Lowest observable adverse effect level
LOEC Lowest observable effect concentration
LOEL Lowest observable effect level
MATC Maximum acceptable toxicant concentration
NAM New approach methodology
NITE National Institute of Technology and Evaluation
NOEC No observable adverse effect concentration
NOAEL No observable effect level
NOEC No observable effect concentration
NOEL No observable effect level
OCSPP Office of Chemical Safety and Pollution Prevention
OPPT Office of Pollution Prevention and Toxics
POD Point of departure
QSAR Quantitative structure-activity relationship (model)
SSD Species sensitivity distribution
TRV Toxicity reference value
TSCA Toxic Substances Control Act
U.S. United States
Web-ICE Web-based Interspecies Correlation Estimation
Page 4 of 54
-------
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
PUBLIC RELEASE DRAFT
December 2024
ACKNOWLEDGEMENTS
This report was developed by the United States Environmental Protection Agency (U.S. EPA or the
Agency), Office of Chemical Safety and Pollution Prevention (OCSPP), Office of Pollution Prevention
and Toxics (OPPT).
Acknowledgements
The Assessment Team gratefully acknowledges the participation, review, and input from EPA OPPT
and OSCPP senior managers and science advisors. The Agency is also grateful for assistance from the
following EPA contractors for the preparation of this draft technical support document: General
Dynamics Information Technology, Inc. (Contract No. HHSN316201200013W); ICF, Inc. (Contract No.
68HERC23D0007); SpecPro Professional Services, LLC (Contract No. 68HERC20D0021); and SRC,
Inc. (Contract No. 68HERH19D0022 and 68HERC23D0007).
As part of an intra-agency review, this technical support document was provided to multiple EPA
Program Offices for review. Comments were submitted by EPA's Office of Research and Development
(ORD).
Docket
Supporting information can be found in the public docket, Docket ID EPA-HQ-QPPT-2018-0434.
Disclaimer
Reference herein to any specific commercial products, process, or service by trade name, trademark,
manufacturer, or otherwise does not constitute or imply its endorsement, recommendation, or favoring
by the United States Government.
Authors: Collin Beachum (Management Lead), Brandall Ingle-Carlson (Assessment Lead), Emily
Griffin (Environmental Hazard Assessment Lead), Jennifer Brennan, Christopher Green (Environmental
Hazard Discipline Leads)
Contributors: Azah Abdallah Mohamed, Rony Arauz Melendez, Sarah Au, Maggie Clark, Jone
Corrales, Daniel DePasquale, Lauren Gates, Ryan Klein, Sydney Nguyen, Brianne Raccor, Maxwell
Sail, Kelley Stanfield, Joe Valdez, Leora Vegosen
Technical Support: Hillary Hollinger, S. Xiah Kragie
This draft technical support document was reviewed and cleared for release by OPPT and OCSPP
leadership.
Page 5 of 54
-------
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
111
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
PUBLIC RELEASE DRAFT
December 2024
SUMMARY
EPA considered all reasonably available information identified through its systematic review process
under the Toxic Substances Control Act (TSCA) to characterize environmental hazard endpoints for
DIBP. Upon evaluating the reasonably available information, environmental hazard thresholds were
derived for aquatic vertebrates, aquatic invertebrates, aquatic benthic invertebrates, aquatic plants and
algae, terrestrial vertebrates, terrestrial invertebrates, and terrestrial plants.
The acute aquatic concentration of concern (COC) for DIBP was derived from a species sensitivity
distribution (SSD) that contained empirical 96-h LC50s for nine species identified in systematic review
as well as an additional 72 species with predicted LC50 and EC50 values from the Web-Based
Interspecies Correlation Estimation (Web-ICE) (v4.0) toxicity value estimation tool (Raimondo. 2010).
The SSD was developed using the The SSD Toolbox (vl. 1), which is a resource created by EPA's
Office of Research and Development (ORD) that can fit SSDs to environmental hazard data (Etterson.
2020). To address data gaps in the DIBP environmental hazard data set, Dibutyl Phthalate (DBP) was
used as an analog and read-across was conducted from the Draft Environmental Hazard Assessment for
Dibutyl Phthalate (DBP) (U.S. EPA. 2024a). Of the nine studies identified in systematic review and
used in the SSD, two studies were from the DIBP empirical data set and seven were from the DBP
empirical data set. The acute COC for aquatic vertebrates, invertebrates, and benthic invertebrates was
identified as 287 |ig/L. All chronic aquatic COCs were calculated using read-across from DBP as an
analog. The chronic aquatic vertebrate COC was identified as 1.56 |ig/L, the aquatic invertebrate COC
was 12.23 |ig/L, the aquatic benthic invertebrate COC was 114.3 mg/kg dry sediment, and the algae
COC was 31.6 |ig/L.
Wildlife mammalian hazard data were not reasonably available; therefore, ecologically relevant
reproductive endpoints from laboratory rodent studies were used to derive a hazard value for terrestrial
mammals. Empirical DIBP toxicity data for rats were used to estimate a hazard value for terrestrial
mammals at 353 mg/kg-bw/day. The terrestrial invertebrate hazard threshold for DIBP was identified as
14 mg DBP/kg dry soil based on read-across from DBP and the terrestrial plant hazard threshold for
DIBP was identified as 10 mg DBP/kg dry soil based on a read-across from DBP (U.S. EPA. 2024a).
EPA's rationale for selecting these hazard thresholds is described in Section 6.
Page 6 of 54
-------
193
194
195
196
197
198
199
200
201
202
203
PUBLIC RELEASE DRAFT
December 2024
1 INTRODUCTION
This technical support document is in support of the Draft Risk Evaluation for Diisobutyl Phthalate
(DIBP) (U.S. EPA. 2025b). Diisobutyl Phthalate (DIBP) is a common name for the chemical substance
1,2-Benzenedicarboxylic acid, l,2-bis(2-methylpropyl) ester (CASRN 84-69-5). See draft risk
evaluation for a complete list of all the technical support documents for DIBP (U.S. EPA. 2025b). DIBP
is an organic substance primarily used as a plasticizer in a wide variety of consumer, commercial and
industrial products. DIBP may be released during industrial activities, manufacturing, disposal, and
through consumer use, with most releases occurring into air and water (U.S. EPA. 2024b). EPA
reviewed studies of the toxicity of DIBP and its analog DBP to aquatic and terrestrial organisms and
DIBP's potential environmental hazards.
Page 7 of 54
-------
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
PUBLIC RELEASE DRAFT
December 2024
2 APPROACH AND METHODOLOGY
During scoping and problem formulation, EPA reviewed potential environmental hazards associated
with DIBP. EPA identified sources of environmental hazard data shown in Figure 2-10 of the Scope of
the Risk Evaluation for DIBP (U.S. EPA. 2020b). EPA completed the review of environmental hazard
data and information sources during risk evaluation using the data quality review evaluation metrics and
the rating criteria described in the 2021 Draft Systematic Review Protocol supporting TSCA Risk
Evaluations for Chemical Substances (U.S. EPA. 2021a) and Draft Risk Evaluation for Diisobiityl
Phthalate (DIBP) - Systematic Review Protocol (U.S. EPA. 2024f). Studies were assigned overall
quality determinations of high, medium, low, or uninformative. EPA systematically evaluated all data
for this hazard characterization but relies upon only high-quality and medium-quality studies for
purposes of risk characterization.
Due to limited environmental hazard data for DIBP, DBP was used as an analog to fill data gaps (U.S.
EPA. 2024a). The criteria for selecting an appropriate analog are structural similarity, similar physical,
chemical, environmental fate and transport behavior in water and sediment, and similar ecotoxicological
behavior in aquatic and benthic taxa based on DIBP toxicity predictions generated using ECOSAR in
comparison to analog (DBP) empirical hazard data. For more information on selecting an analog, see
Appendix A.
An SSD analysis was conducted using EPA's SSD Toolbox (vl.l) to determine an acute aquatic hazard
threshold. A SSD is a type of probability distribution of toxicity values from multiple species. It can be
used to visualize which species are most sensitive to a toxic chemical exposure, and to predict a
concentration of a toxic chemical that is hazardous to a percentage of test species. Predicted hazard data
were generated using EPA's Web-ICE (v4.0) toxicity predictions tool (Raimondo. 2010). Empirical data
that were included in the SSD analysis were limited to at or below the limit of water solubility of 6.2
mg/L for DIBP (U.S. EPA. 2024d). The specific species and corresponding empirical data are outlined
in Section 3 and a description on the SSD as well as values predicted through EPA's Web-ICE tool can
be found in Appendix B.
Page 8 of 54
-------
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
PUBLIC RELEASE DRAFT
December 2024
3 AQUATIC SPECIES HAZARD
EPA reviewed a total of three studies for DIBP toxicity to aquatic organisms and 171 studies for DBP.
Of these studies, those that received an overall quality determination of low or uninformative were not
considered for quantitative risk evaluation. Further, studies that received an overall quality
determination of high and medium, but demonstrated no acute or chronic adverse effects at the highest
concentration tested (unbounded no-observed-effect-concentration [NOECs]), or where hazard values
exceeded the limit of solubility for DIBP in water as determined by EPA at 6.2 mg/L (U.S. EPA.
2024d), were excluded from consideration for development of hazard thresholds. Therefore, for DIBP,
two studies were considered for the development of hazard thresholds as one aquatic algae study
received an overall quality determination of low (described below). For all DBP studies that were
excluded from the quantitative analysis, please see Appendix C of the Draft Environmental Hazard
Assessment for Di butyl Phthalate (DBP) (U.S. EPA. 2024a). For the analog DBP, the hazard values
from studies which were used to derive hazard thresholds were used as read-across for DIBP and are
described below (Table 3-1). These hazard values were the most sensitive, had clear population-level
fitness endpoints and were selected as the most appropriate in the DBP data set to represent hazard. For
all data considered for the DBP risk evaluation see the Draft Environmental Hazard Assessment for
DibutylPhthalate (DBP) (U.S. EPA. 2024a).
Studies that received an overall quality determination of uninformative were not considered or included
in the quantitative risk assessment. Additionally, studies that received an overall quality determination
of low were supplemented with read across. EPA identified 21 aquatic toxicity studies (two DIBP
studies and 19 DBP studies). The DIBP acute aquatic and benthic hazard data along with the acute DBP
analog data described below were used to generate Web-ICE toxicity predictions for additional taxa
representation. Specifically, predicted hazard data for 72 species were generated using EPA's Web-ICE
tool, including predictions for 39 fish species, 31 invertebrate species, and 2 amphibian species
(Table Apx B-2)Empirical and predicted hazard values were used as input in an SSD analysis to
determine an acute aquatic hazard threshold.
Toxicity in Aquatic Vertebrates
One acute aquatic vertebrate study was available for the quantitative assessment of potential hazards
from DIBP exposure. An additional six studies with empirical acute aquatic DBP data were used as an
analog for DIBP (Table 3-1). For DIBP acute aquatic vertebrates, EPA conducted a study in which
fathead minnows were exposed to several phthalates, including DIBP and DBP, for 24 hours (Bencic et
al.. 2024) and a 24-hour mortality LC50 of 5.6 mg/L was identified for DIBP (Table 3-1). The additional
six studies with analog DBP represent three species of aquatic vertebrates with six hazard values. In
bluegill {Lepomis macrochirus), the 96-hour mortality LC50s for aquatic DBP exposure ranged from
0.48 to 1.2 mg/L (Adams et al.. 1995; EG&G Bionomics. 1983b; Buccafusco et al.. 1981). In rainbow
trout (Oncorhynchus mykiss), the 96-hour mortality LC50s for aquatic DBP exposure ranged from 1.40
to 1.60 mg/L (EnviroSvstem. 1991). In zebrafish (Danio rerio), a 72-hour mortality LC50 of 0.63 mg/L
DBP was identified (Chen et al.. 2014).
TSCA Section 4(h)(1)(B) requires EPA to encourage and facilitate the use of scientifically valid test
methods and strategies that reduce or replace the use of vertebrate animals while providing information
of equivalent or better scientific quality and relevance that will support regulatory decisions. In line with
EPA's New Approach Methods Work Plan, EPA OPPT and ORD have been collaborating on
developing new methods for use in TSCA risk evaluations. Specifically, a project was conducted to
generate omics-based PODs and compared them to traditional endpoints using fathead minnow as the
model organism for three of the phthalates undergoing a TSCA risk evaluation, including DIBP (Bencic
et al.. 2024). In this study, points of Departure (PODs) were derived for transcriptomic change (tPOD;
Page 9 of 54
-------
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
PUBLIC RELEASE DRAFT
December 2024
0.87 mg/L), metabolomic change (mPOD; 0.15 mg/L), and behavioral change (bPOD 0.90 mg/L)
resulting from 24 hour duration of aquatic DIBP exposure to fathead minnows. These results suggest
that fathead minnow larvae exhibited changes in gene expression, metabolite levels, and swimming
behavior at sublethal concentrations of DIBP. While hazard thresholds are usually calculated with in
vivo data measuring an apical endpoint (e.g., mortality, reproduction, growth), these mechanistic
(transcriptomic and metabolomic) and behavior points of departure represent potential information that
may be used for reducing the time needed for toxicity testing in vivo and provide an alternate method to
characterize hazard as well as provide important evidence for mechanisms of action. At this time, EPA
has not used the omics-based PODs in the DIBP draft risk evaluation. There are uncertainties with
respect to the extent to which these sub-organismal and individual-level effects (e.g., behavior) at short
exposure durations are comparable to ecologically relevant outcomes, such as survival and reproduction,
in wild fish populations.
No chronic aquatic vertebrate studies were available for the quantitative assessment of potential hazards
from DIBP exposure. Therefore, a read-across was conducted using the hazard value used to derive a
hazard threshold identified from the DBP data set as an analog for chronic aquatic vertebrate hazard
data. From the DBP hazard data set, 11 studies with overall quality determinations of high and medium
contained chronic endpoints that identified definitive hazard values for five fish species and two
amphibians (U.S. EPA. 2024d). The hazard threshold identified in DBP resulted from a
multigenerational Japanese medaka (Oryzias latipes) study in which parental fish were aqueously
exposed to DBP at measured concentrations of 15.6, 38.7, 66, 103, and 305 |ig/L. Significant effects
were observed in growth of both male and female F1 and F2 generations. In the male and female F1
generations, weight was significantly less compared to controls at 112-days, resulting in no observed
effect concentrations/lowest observed effect concentration (NOECs/LOECs) of <15.6/15.6 |ig/L and
66/103 |ig/L DBP in males and females, respectively. Additionally, in the F2 generation, weight was
significantly lower compared to controls at day 98, resulting in NOECs/LOECs of 103/305 |ig/L and
15.6/38.7 |ig/L DBP in males and females, respectively (EAG Laboratories. 2018). Unbounded effects
(unbounded LOEC) were also observed for growth at the lowest concentration tested. Specifically, male
F1 adult weight at 112-days, male F2 adult weight and length at 70-days, and male F2 adult length at 98-
days were significantly inhibited at 0.015 mg/L DBP. The LOEC of 15.6 |ig/L DBP for a reduced
weight in the male F1 generation was chosen for COC calculations.
Toxicity in Aquatic Invertebrates
No acute or chronic aquatic invertebrate studies were available for the quantitative assessment of
potential hazards from DIBP exposure. Therefore, a read-across was conducted using the acute and
chronic hazard thresholds identified for aquatic invertebrates exposed to DBP. From the acute DBP
hazard data set, four studies with hazard data for two aquatic invertebrate species were included in the
SSD analysis for DIBP. In the opposum shrimp (Americamysis bahia), the mortality 96-hour LC50s
ranged from 0.50 to 0.75 mg/L (Adams et al.. 1995; EG&G Bionomics. 1984a). In the water flea
(Daphnia magna), the 48-hour mortality LC50s ranged from 2.55 to 5.2 mg/L (Wei et al.. 2018;
McCarthy and Whitmore. 1985). From the DBP chronic hazard data set, eight studies contained
endpoints that identified definitive hazard values for 10 aquatic invertebrate species. The hazard value
chosen to derive a hazard threshold for chronic invertebrates resulted from a 14-day study of the
Amphipod crustacean (Monocorophium acherusicum), which were maintained in measured aqueous
concentrations of 0.044, 0.34, and 3.7 mg/L DBP. An observed 90 percent reduction in abundance was
observed at 0.34 mg/L DBP resulting in 14-day NOEC/LOEC of 0.044/0.34 mg/L and a chronic value
or geometric mean of the NOEC/LOEC (ChV) of 0.112 mg/L (Tagatz et al.. 1983).
Toxicity in Aquatic Benthic Invertebrates
Page 10 of 54
-------
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
PUBLIC RELEASE DRAFT
December 2024
Acute invertebrate hazard data for DIBP was identified in one medium-rated study representing a 96-
hour exposure to the harpacticoid copepod (Nitocra spinipes). The static 96-hour LC50 for mortality
was measured at 3 mg/L DIBP (Linden et al.. 1979) (Table 3-1). No additional acute and no chronic
benthic invertebrate studies were available for the quantitative assessment of potential hazards from
DIBP exposure. Therefore, a read-across was conducted using acute hazard studies for benthic
invertebrates exposed to DBP as well as a read-across of the hazard value chosen to derive a hazard
threshold from the DBP data set as an analog for chronic aquatic benthic invertebrate hazard. In the
midge (Paratcmytarsusparthenogeneticus) and the midge (Chironomusplumosus) DBP acute benthic
invertebrate hazard data set, the 48-hour mortality LC50s ranged from 4.0 to 5.8 mg/L DBP (EG&G
Bionomics. 1984b; Streufort 1978). Acute aquatic hazard values were included in the SSD analysis.
From the DBP chronic benthic invertebrate hazard data set, the hazard threshold was identified from
Call et al. (2001). which studied the effects of DBP in pore water and sediment for high, medium, and
low TOC (total organic carbon) in the midge (Chironomus tentcms). For high TOC, a 10-day
NOEC/LOEC of 0.448/5.85 mg/L DBP in pore water and 508/3550 mg/kg dry weight DBP in sediment
was observed for an increase in weight. For medium TOC, a 10-day NOEC/LOEC of 3.85/16 mg/L
DBP in pore water and 423/3090 mg/kg dry weight DBP in sediment was observed for an increase in
weight. For mortality, the 10-day NOEC/LOEC for pore water and sediment in high, medium, and low
TOC was 0.448/5.85 mg/L DBP and 508/3550 mg/kg dry weight DBP, 3.85/16 mg/L DBP and
423/3090 mg/kg dry weight DBP, and 0.672/4.59 mg/L DBP and 50.1/315 mg/kg dry weight DBP,
respectively (Call et al.. 2001). The data resulting from the medium TOC sediment group was chosen to
derive a COC as this is the closest to the assumed TOC level (four percent) used in Point Source
Calculator (EPA. 2019) to estimate DIBP exposure in benthic organisms.
Toxicity in Amphibians
No amphibian studies were available for the quantitative assessment of potential hazards from DIBP
exposure. Web-ICE predictions generated using both DIBP and DBP acute aquatic hazard data
identified four 24-hour LC50s for two amphibian species. In the bullfrog (.Lithobates catesbeicimis), a
24-hour LC50 of 2.98 mg/L (a geometric mean of three predicted values for the same species) was
predicted. In the African clawed frog (Xenopus laevis), a 24-hour LC50 of 4.0 mg/L was predicted.
These data were used in the acute SSD analysis.
Toxicity in Aquatic Plants
One low-quality study was available for the assessment of potential hazards from DIBP exposure to
aquatic algae. In this study, no effects were observed on population growth in algae (Karenia brevis)
exposed to 0 to 200 ml/L DIBP for seven days (Liu et al.. 2016). Since EPA relies upon only high-
quality and medium-quality studies for purposes of quantitative risk characterization, a read-across was
conducted using the hazard value chosen to derive a hazard threshold from the DBP data set as an
analog for aquatic plant and algae hazard data. The DBP hazard data set contained three high- or
medium-rated studies with endpoints that identified definitive hazard values for one species of algae
(U.S. EPA. 2024d). The hazard value used to derive a hazard threshold for DBP resulted from a
medium-quality green algae (Selenastrum capricornatam) study (Adachi et al.. 2006). which identified a
96-hour NOEC/LOEC of 0.1/1.0 mg/L in S. capri cornatam at DBP measured concentrations ranging
from 0.1 to 10 mg/L (Adachi et al.. 2006).
Table 3-1. Aquatic Organisms Environmental Hazard Studies Used for DIBP, Supplemented with
Page 11 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
377 DBP Environmental Hazard Data
Test Organism
Hazard
Values
Duration
Phthalate
Endpoint
Citation
(Study
Quality)
Aquatic Vertebrates
Acute
Fathead minnow
(Pimephales
promelas)
5.3 mg/L°
24-hr
LC50
DIBP
Mortality
(Bencic et al.,
2024)(High)
Acute
Bluegill {Lepomis
macrochirus)
1.2 mg/L°
96-hr
LC50
DBP
Mortality
(Buccafusco et
al.. 1981)
(Medium)
0.85 mg/L°
96-hr
LC50
DBP
Mortality
(EG&G
Bionomics,
1983b)(Hiah)
0.48 mg/L°
96-hr
LC50
DBP
Mortality
(Adams et al..
1995)(High)
Rainbow trout
(Oncorynchus
mykiss)
1.60 mg/L°
96-hr
LC50
DBP
Mortality
(EG&G
Bionomics,
1983a) (Hiah)
1.40 mg/L°
96-hr
LC50
DBP
Mortality
(EnviroSvstem,
1991)(High)
Zebrafish (Danio
rerio)
0.63 mg/L
72-hr
LC50
DBP
Mortality
(Chen et al.,
2014)
(Medium)
Chronic
Japanese medaka
{Oryzias latipes)
<15.6/15.6
^g/L
112-d
NOEC/
LOEC
(ChV)
DBP
Growth -
Weight male
F1 Adults
(EAG
Laboratories,
2018)(High)
Aquatic Invertebrates
Acute
Opossum shrimp
(Americamysis
bahia)
0.75 mg/L°
96-hr
LC50
DBP
Mortality
(EG&G
Bionomics,
1984a) (Hiah)
0.50 mg/L°
96-hr
LC50
DBP
Mortality
(Adams et al.,
1995)(High)
Water flea
{Daphnia magna)
5.2 mg/L°
48-hr
LC50
DBP
Mortality
(McCarthy and
Whitmore,
1985)
(Medium)
2.55 mg/L°
48-hr
LC50
DBP
Mortality
(Wei et al.,
2018)(High)
Page 12 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Test Organism
Hazard
Values
Duration
Phthalate
Endpoint
Citation
(Study
Quality)
4.31 mg/L°
48-hr
LC50
DBP
Mortality
2.83 mg/L°
48-hr
LC50
DBP
Mortality
Chronic
Amphipod
crustacean
(Monocorophium
acherusicum)
0.044/0.34
mg/L (0.122
mg/L)
14-d
NOEC/
LOEC
(ChV)
DBP
Population -
Abundance
(Tasatz et al
1983)
(Medium)
Aquatic
3enthic Invertebrates
Acute
Harpacticoid
copepod (Nitocra
spinipes)
3 mg/L°
96-hr
LC50
DIBP
Mortality
(Linden et al
1979)
(Medium)
Harpacticoid
copepod (Nitocra
spinipes)
1.7 mg/L°
96-hr
LC50
DBP
Mortality
(Linden et al
1979)
(Medium)
Midge
(Paratany tarsus
parthenogeneticus)
5.8 mg/L°
48-hr
LC50
DBP
Mortality
(EG&G
Bionomics,
1984b)(Hi ah)
Midge
(Chironomus
plumosus)
4.0 mg/L°
48-hr
LC50
DBP
Mortality
(Streufort,
1978)
(Medium)
Chronic
Midge
(Chironomus
tentans)
423/3090
mg/kg
(1143
mg/kg) dry
weight
10-d
NOEC/
LOEC
(ChV)
DBP
Mortality
(Call et al
2001)(High)
Aquatic Plants anc
Algae
Green algae
(Selenastrum
capri cornutum)
0.1/1 mg/L
96-h
NOEC/
LOEC
DBP
Population
(Abundance)
(Adachi et al..
2006)
(Medium)
'Value used in SSD analysis and used to inform web-ice predictions. Water solubility of DB
water solubility of DIBP = 6.2 mg/L.
5 = 11.2 mg/L and
378
Page 13 of 54
-------
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
PUBLIC RELEASE DRAFT
December 2024
4 TERRESTRIAL SPECIES HAZARD
Two wildlife terrestrial studies were identified for DIBP, one with a quality determination of high and
one with a quality determination of medium. These studies contained relevant toxicity data for the
nematode (Caenorhabditis e legem s) and the tobacco plant (Nicotiana tabacum). Additionally, in lieu of
wild terrestrial mammal studies, two references for human health model organisms (Sprague-Dawley
rats, Rattus norvegicus) were used to determine terrestrial vertebrate hazard values. These studies were
used to determine the lowest and thus most conservative DIBP concentration that displayed apical
endpoint effects (e.g., survival, reproduction, growth) in rodents, and which could also serve as
representative of hazard effects in wild mammal populations. These dietary DIBP concentrations were
expressed as doses in mg/kg bw/day, and since body weight was normalized, EPA used this data as a
screening surrogate for the effects on ecologically relevant wildlife species to evaluate chronic dietary
exposure to DIBP. One high-quality study on the springtail (Folsomia fimetaria) and a high-quality
study on bread wheat (Triticum aestivum) were also included to fill data gaps in the DIBP data set
Terrestrial species hazard data are displayed in Table 4-1, as the most relevant for quantitative
assessment.
Toxicity in Terrestrial Vertebrates
EPA reviewed two laboratory rodent studies from human health animal models for hazards of DIBP as
surrogates to wild mammal populations, which contained ecologically relevant reproductive endpoints
with both a no observed effect level (NOAEL) and lowest observed effect level (LOAEL) represented
for each endpoint (Saillenfait et al.. 2008; Saillenfait et al.. 2006). EPA's decision to focus on
ecologically relevant (population level) reproductive endpoints in the rat and mouse data set for DIBP
for consideration of a hazard threshold in terrestrial mammals is due to the known sensitivity of these
taxa to DIBP in eliciting phthalate syndrome (U.S. EPA. 2025a). EPA focused on studies which
contained both a NOAEL and a LOAEL for each reproductive endpoint to refine the hazard threshold.
Of the two rat studies containing NOAEL-LOAEL pairs for ecologically relevant reproductive
endpoints, EPA selected the study with the most sensitive, and thus most conservative, LOAEL for
deriving the hazard threshold for terrestrial mammals. In one study, pregnant Sprague-Dawley rats were
given DIBP at doses of 0 (olive oil), 250, 500, 750, and 1000 mg/kg/day for 21 days via gavage. A
significant decrease in maternal body weight gain was observed starting at gestational days six through
nine at concentrations greater than 500 mg/kg/day and the percent of resorptions per litter was
significant at 750 mg/kg/day (27.6 percent). In both male and female fetuses, body weight was
significntly lower (9 percent) at 500 mg/kg/day compared to controls, resulting in a gestational day 20
NOAEC/LOAEC of 250/500 mg/kg/day (Saillenfait et al.. 2006). This study was used for hazard value
calculations. In the other study, pregnant Sprague-Dawley rats were given DIBP on gestation days 6-21
at doses of 0 (olive oil), 125, 250, 500, and 625 mg/kg/day via gavage. No effects were observed at any
dose in pregnant females nor were any effects observed on litter size. However, at 500 and 625
mg/kg/day, male pup weight was lower than controls by six to eight percent and 10 to 12 percent,
respectively. Additionally, male and female pup weight was significantly less than the control on post-
natal day (PND) 1 at 625 mg/kg/day (Saillenfait et al.. 2008).
Toxicity in Terrestrial Invertebrates
Acute terrestrial invertebrate hazard data for DIBP was identified in one high-ranking study. Nematodes
maintained in culture media with DIBP for 24 hours at nominal concentrations of 0, 100, and 1000 mg/L
DIBP were observed to have significant effects on behavior at 100 mg/L. Specifically, nematodes
exhibited changes in distance moved, reversals, and overall movements at the lowest concentrations
tested compared to controls (Tseng et al.. 2013). However, this study only tested concentrations that
exceeded the DIBP limit of water solubility (6.2 mg/L), therefore a read-across was conducted from the
DBP data set. The DBP hazard data set contained 12 high- or medium-rated studies with definitive
Page 14 of 54
-------
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
PUBLIC RELEASE DRAFT
December 2024
endpoints that identified hazard values for seven terrestrial invertebrate species (U.S. EPA. 2024d). The
hazard value used to derive a hazard threshold for DBP from was from a high-ranking study that
examined the effects of DBP in the springtail (Folsomia fimetaria). In this study, adult springtail
reproduction was significantly affected with an observed 21-day EC10 and EC50 of 14 and 68 mg/kg
dry soil, respectively (Jensen et al.. 2001).
Toxicity in Terrestrial Plants
One medium-ranking study was available to assess DIBP toxicity to terrestrial plants. The toxicity of
DIBP to two tobacco plant (Nicotiana tabacum) seed cultivars, G168 and Hong da, was assessed using
filter paper at nominal concentrations of 0.1, 0.5, 1.0, and 10 mM (0, 27.8, 139, 278, and 2,783 mg/L)
DIBP. In the G168 cultivar, seed germination was significantly reduced (28 percent germination) at the
highest concentration tested and in the Hong da cultivars, seed germination was significantly reduced
(44 percent germination) at 0.5 mM. Thus, the 7-day NOEL/LOEL for seed germination was found to be
1.0/10 mM (278/2,783 mg/L) for G168 cultivars, and 0.1/0.5 mM (27.8/139 mg/L) for Hong da cultivars
(Jia et al.. 2011). However, this study only tested concentrations that exceeded the DIBP limit of water
solubility (6.2 mg/L), therefore this data was not used in the quantitative assessment of DIBP hazards
and read across was conducted from DBP for terrestrial plants. In bread wheat exposed to DBP at
concentrations of 0, 5, 10, 20, 30, and 40 mg/L, a 40-day LOEL of 10 mg/kg DBP (lowest concentration
used in the study) for reduced weight in bread wheat was observed (Gao et al.. 2019).
Table 4-1. Terres
trial Organisms Environmental Hazard Studies Used for I
HBP
Test Organism
Hazard
Values
Duration
Phthalate
Endpoint
Citation
(Study Quality)
Terrestrial Vertebrates
Sprague-
Dawley rat
250/500
mg/kg/day"
Gestational day
20 NOAEL/
LOAEL
DIBP
Reproduction
(Saillenfait et al.,
2006)(High)
Sprague-
Dawley rat
250/500
mg/kg/day
Gestational day
21 NOAEL/
LOEL
DIBP
Reproduction
(Saillenfait et al.,
2008)(High)
Terrestrial Invertebrates
Nematode
(Caenorhabditis
elegans)
<100/100
mg/L
(culture
media)6
24-hr NOEL/
LOEL
DIBP
Behavior
(Tsens et al., 2013)
(High)
Springtail
(Folsomia
fimetaria)
14 mg/kg
dry soil0
21-d EC10
DBP
Reproduction
(Jensen et al., 2001)
(High)
Terrestrial Plants
Tobacco
(Nicotiana
tabacum) G168
cultivar
278/2,283
mg/L6
7-d NOEL/ LOE1
DIBP
Reproduction
- germination
(Jia et al., 2011)
(Medium)
Tobacco
(Nicotiana
tabacum) Hong
da cultivars
27.8/139
mg/L6
DIBP
Bread wheat
(Triticum
<10 mg/kg
dry soil/10
40-day LOEL
DBP
Growth
(Gao et al., 2019)
(High)
Page 15 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Test Organism
Hazard
Values
Duration
Phthalate
Endpoint
Citation
(Study Quality)
aestivum)
mg/kg dry
soil0
l7Value used to derive a hazard value; ''Value exceeds the DIBP limit of water solubility (6.2 mg/L)
449
450
Page 16 of 54
-------
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
PUBLIC RELEASE DRAFT
December 2024
5 WEIGHT OF SCIENTIFIC EVIDENCE CONCLUSIONS FOR
ENVIRONMENTAL HAZARD ASSESSMENT
EPA determined that DIBP poses potential hazard to acute aquatic species at aquatic concentrations of
287 |ig/L, as determined through SSD supplemented with DIBP empirical data, DBP empirical data, and
predicted values calculated through Web-ICE. EPA determined that DIBP poses potential chronic
hazard effects to aquatic species based on read-across conducted from DBPj (U.S. EPA. 2024a). which
evaluated studies on DBP chronic toxicity in aquatic vertebrates, invertebrates, and benthic invertebrates
as an analog to DIBP. The endpoints used in the read-across were the hazard values used to derive
hazard thresholds in DBP, which were the most sensitive, clear population-level fitness endpoints
selected as the most appropriate in the DBP data set to represent hazard. For all studies considered in the
DBP hazard assessment, see the Draft Environmental Hazard Assessment for Dibutyl Phthalate (DBP)
(U.S. EPA. 2024a).
EPA determined that DIBP poses potential hazards to terrestrial mammals at a dietary dose of 353
mg/kg/day, which is supported by evidence taken from laboratory rodent studies used as human health
models (Saillenfait et al.. 2006). EPA determined DIBP poses potential hazards to terrestrial
invertebrates based on read-across from DBP (U.S. EPA. 2024a). in which a hazard value of 14 mg/kg
dry soil was identified (Jensen et al.. 2001). EPA determined DIBP poses potential hazards to terrestrial
plants based on read-across from DBP (U.S. EPA. 2024a). in which a hazard value of 10 mg/kg dry soil
was identified (Gao et al.. 2019).
The aquatic COCs and terrestrial hazard thresholds identified in this technical support document will be
used in the Draft Risk Evaluation for Diisobiityl Phthalate (DIBP) (U.S. EPA. 2025b) to characterize
environmental risk.
5.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty
for the Environmental Hazard Assessment
EPA has robust confidence that DIBP poses potential hazard to acute aquatic species at 287 (J,g/L. This
data is supported through SSD analysis conducted with empirical data from two acute DIBP aquatic
hazard studies, seven acute DBP aquatic hazard values, and supplemented with predicted values
calculated through Web-ICE. A limitation and source of uncertainty in the assessment of hazards to
chronic aquatic organisms is the lack of available data. No aquatic chronic studies were available for the
quantitative assessment of potential hazards from DIBP exposure. Therefore, a read-across was
conducted from DBP (U.S. EPA. 2024a). DBP was considered an appropriate analog for DIBP based on
structural similarity, similar physical, chemical, environmental fate and transport behavior in water and
sediment, as well as similar ecotoxicological behavior in aquatic taxa (Appendix A). EPA has robust
confidence that DIBP poses hazard to aquatic vertebrates, invertebrates, and benthic invertebrates on a
chronic basis. This robust confidence is supported by the quality and consistency of the analog DBP
chronic aquatic vertebrate, invertebrate, and benthic invertebrate database. A read-across from DBP was
also conducted for aquatic plants and algae. However, only one species of algae was available for the
assessment of potential hazards from DBP (U.S. EPA. 2024a). therefore EPA has overall moderate
confidence in the hazard for the DIBP aquatic plants and algae assessment. For more information on
analog selection, see Appendix A.
In the terrestrial environment, EPA has moderate confidence that DIBP poses potential hazard to
mammals, and robust confidence that DIBP poses potential hazard to invertebrates, and plants. The
conclusion that DIBP poses hazard to terrestrial mammals at a dietary dose of 353 mg/kg/day, is
supported by evidence obtained from laboratory rodent studies used as human health models. Utilizing
Page 17 of 54
-------
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
PUBLIC RELEASE DRAFT
December 2024
human health rodent models as a surrogate for terrestrial models introduces uncertainty into the
terrestrial hazard characterization since these species may not fully represent effects observed in wild
animal populations. The conclusion that DIBP poses hazard to terrestrial invertebrates is based on one
study that identified significant behavioral effects in the nematode (Tseng et al.. 2013). A limitation and
uncertainty of the terrestrial invertebrate data set is the low number of available studies and species
available to be used in the assessment. However, the strength of the database and identified hazard value
is supported by the robust consistency, strength and precision, and biological gradient of the study
results. EPA has moderate confidence that DIBP poses hazard to terrestrial plants. This confidence is
supported by the quality and consistency of the analog DBP terrestrial plant database. Due to the added
uncertainty from some studies in similar plants showing a lack of strong biologically relevant effects or
clear dose-response, confidence was reduced for the strength and precision and dose-response
considerations for the terrestrial plants assessment.
5.1.1 Confidence in the Environmental Hazard Data set
Based on the weight of the scientific evidence and uncertainties, a confidence statement was
developed that qualitatively ranks {i.e., robust, moderate, slight, or indeterminate) the confidence
in the hazard threshold. The evidence considerations and criteria detailed within the Draft
Systematic Review Protocol (U.S. EPA, 2021a) guide the application of strength-of-evidence
judgments for environmental hazard effect within a given evidence stream. See Appendix C for
more information on the weight of scientific evidence conclusions and see
Table 5-1 for the confidence table that summarizes the information below.
For the acute aquatic assessment of DIBP, the database consisted of two studies, one with an overall
quality determination of medium and another conducted by EPA with an overall quality determination
of high. These two studies, plus data from seven additional studies from the Draft Environmental
Hazard Assessment for Dibutyl Phthalate (DBP) (U.S. EPA. 2024a). as well as 72 hazard endpoints
obtained from Web-ICE predictions were used to generate an SSD output. Thus, for the acute data set, a
robust confidence was assigned to the quality of the database. The studies from the analog DBP data set
displayed similar effects on the same species across multiple studies and these effects were similar to
what was observed in the two acute DIBP studies. Due to the observed consistent effects, a robust
confidence was assigned to the consistency consideration for the acute aquatic assessment. The effects
observed in both the acute DBP and DIBP data set were apical endpoints such as 48-hour, 72-hour, or
96-hour LC50s with additional predicted LC50 values reported from Web-ICE. Therefore, a robust
confidence was assigned to the strength and precision consideration. As dose-response is a prerequisite
of obtaining reliable LC50 values and was observed in the empirical studies that were used in the SSD, a
robust confidence was assigned to the dose-response consideration. Lastly, for the acute aquatic
assessment, mortality was observed in the empirical data for four fish and five invertebrates and
mortality was predicted in 72 additional species using Web-ICE. The use of the lower 95 percent
confidence interval (CI) of the 5th percentile hazardous concentration (HC05) in the SSD instead of a
fixed assessment factor (AF) also increases confidence since it is a more data-driven way of accounting
for uncertainty. Due to the use of empirical data combined with predicted data through a probabilistic
approach, a robust confidence was assigned to the relevance consideration for the acute aquatic
assessment.
No studies were available for the chronic aquatic vertebrate assessment of DIBP. Therefore, a read-
across was conducted from DBP (U.S. EPA. 2024a). Eleven studies from the analog DBP contained
chronic endpoints that identified definitive hazard values below the DIBP limit of water solubility for
aquatic vertebrates (6.2 mg/L), (U.S. EPA. 2024d) for five fish species and two amphibians, resulting in
robust confidence for quality of the database. DBP displayed chronic effects on growth which spanned
Page 18 of 54
-------
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
PUBLIC RELEASE DRAFT
December 2024
several orders of magnitude among aquatic vertebrate taxa, therefore a moderate confidence was
assigned to the consistency of the database. In the study chosen to derive the COC, (EAG Laboratories.
2018). body weight in Japanese medaka was inhibited by 13.4 percent relative to the vehicle control, and
there was a statistically significant trend toward greater body weight inhibition with increasing dose,
culminating at 34.0 percent inhibition at the highest dose (305 |ig/L). Strong dose-response effects were
also observed in other studies in the DBP database. Therefore, a robust confidence was assigned to the
strength and precision consideration and the dose-response consideration for the chronic aquatic
invertebrate assessment. Lastly, due to ecologically relevant population level effects (growth and
mortality) observed in multiple species for DBP, yet the data being represented by an analog, a moderate
confidence was assigned to the relevance consideration for the chronic aquatic vertebrate assessment.
All studies considered for DBP can be found in the Draft Environmental Hazard Assessment for Dibutyl
Phthalate (DBP) (U.S. EPA. 2024a).
No studies were available for the chronic aquatic invertebrate assessment of DIBP. Therefore, a read-
across was conducted from DBP (U.S. EPA. 2024a). Eight studies from the analog DBP contained
chronic endpoints that identified definitive hazard values below the DIBP limit of water solubility for 10
aquatic invertebrate species, resulting in robust confidence for quality of the database. The studies from
DBP database had similar effects on the same species across multiple studies, and within one order of
magnitude. Therefore, a robust confidence was assigned to the consistency consideration. In the study
chosen to derive the COC (Tagatz et al.. 1983). amphipod populations were reduced by 91 percent at the
LOEC and 100 percent mortality was observed at higher doses. A strong dose-response relationship was
also observed in the other studies from the analog DBP database and therefore a robust confidence was
assigned to strength and precision and dose-response of the database for the chronic aquatic invertebrate
assessment. For the chronic aquatic invertebrate assessment, ecologically relevant population level
effects (mortality and reproduction) were observed in 10 species, two of which (water flea, Daphnia
magna; and the worm Lumbricuius variegatus) are considered representative test species for aquatic
toxicity tests. Similarly to the chronic aquatic vertebrate assessment, ecologically relevant population
level effects were observed in multiple species for DBP, yet the data was represented by an analog,
therefore a moderate confidence was assigned to the relevance consideration for the chronic aquatic
invertebrate assessment. All studies considered for DBP can be found in the Draft Environmental
Hazard Assessment for Dibutyl Phthalate (DBP) (U.S. EPA. 2024a).
No studies were available for the chronic aquatic benthic invertebrate assessment of DIBP. Therefore, a
read-across was conducted from DBP (U.S. EPA. 2024a). Three studies from the analog DBP contained
chronic endpoints that identified definitive hazard values below the DIBP limit of water solubility for
benthic invertebrates (U.S. EPA. 2024d). These studies included multiple species, endpoints, and
durations, however only two species were represented. Additionally, the results seemed to be repeated
across some of the studies and it was unclear in some cases whether the data were original. These
considerations resulted in a slight confidence assigned for the quality of the database consideration. DBP
studies were conducted with low, medium, and high TOC sediments. Among the same species, effects
were generally within one order of magnitude in the same TOC. Therefore, a robust confidence was
assigned to the consistency of the database. In the study chosen to derive the COC (Lake Superior
Research Institute. 1997). the midge population was reduced by 76.7 percent at the LOEC (3,090 mg
DBP/kg dry sediment) and population reduction in other treatments and TOC levels was consistent.
Therefore, a robust confidence was assigned to the strength and precision of the database. In the medium
TOC group, higher doses of DBP displayed similar mortality. Due to a clear dose-response relationship
in other studies in the database, a moderate confidence was assigned to the dose-response consideration
for the chronic benthic invertebrate assessment. Ecologically relevant population level effects were
observed in two different species from the DBP database (scud, Hyalella azteca\ and midge,
Page 19 of 54
-------
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
PUBLIC RELEASE DRAFT
December 2024
Chironomus plumosus), both of which are considered representative test species for benthic toxicity
tests. However, relevance is limited by the use of an analog, therefore, moderate confidence was
assigned to the relevance consideration for the chronic benthic invertebrate assessment.
No studies were available for the aquatic plant or algae assessment of DIBP. Therefore, a read-across
was conducted from DBP (U.S. EPA. 2024a). DBP database consisted of seven high or medium quality
studies for toxicity in aquatic plants and algae. Three studies from the analog DBP contained endpoints
that identified definitive hazard values below the DIBP limit of water solubility (U.S. EPA. 2024d) for
one species of green algae. Confidence in the database was reduced because only one species was
identified and several of the studies in the database were not acceptable due to exposure concentrations
being above the limit of solubility for DIBP, therefore a slight confidence was assigned for the quality of
the database. DBP had similar effects on population, measured as either chlorophyll a concentration or
cell abundance, in three independent studies. Thus, a robust confidence was assigned to the consistency
of the database. In the study chosen to derive the COC (Adachi et al.. 2006). a significant reduction in
the algal population was observed at the LOEC (1000 |ig/L DBP) and population reduction was
increased with higher concentrations of DBP. However, there was an increase in algal population at the
NOEC (100 |ig/L DBP), therefore a moderate confidence was assigned to the strength and precision and
dose-response considerations for the aquatic plants and algae assessment. An ecologically relevant
population level effect (population abundance, measured as either chlorophyll a concentration or cell
count) was observed in one species of green algae (Selenastrum capricornutum). Due to this species
being considered a representative test species for algal toxicity tests, yet being the only species
represented in the database and the use of an analog, moderate confidence was assigned to the relevance
consideration for the aquatic plant and algae assessment.
For the terrestrial vertebrate assessment, EPA reviewed two laboratory rodent studies as surrogates from
human health animal models for hazards of DIBP to wild mammal populations (Saillenfait et al.. 2008;
Saillenfait et al.. 2006). While two terrestrial vertebrate studies were available for the assessment of
DIBP, these studies were not from wildlife species and therefore a moderate confidence was assigned to
the quality of the database. In these studies, effects on growth and reproduction were observed at
NOAEL/LOAELs ranging from 250/500 mg/kg/day from to 500/750 mg/kg/day DIBP (Saillenfait et al..
2008; Saillenfait et al.. 2006). Since significant effects occurred at similar doses and concentrations
across studies, a robust confidence was assigned to the consistency of the database. In the study chosen
to derive a terrestrial vertebrate hazard value, a significant reduction (seven percent) in body weight for
both male and female fetuses resulting in a gestational day 20 NOEC/LOEC of 250/500 mg/kg/day was
observed (Saillenfait et al.. 2006). Body weight was also significantly reduced at the higher
concentrations of 750 and 1000 mg/kg/day by 17 percent and 24 percent, respectively. Similar dose-
response relationships were also observed for the other endpoints in the study. Thus, a robust confidence
was assigned for the dose-response and strength and precision of the database considerations. Data from
human-relevant terrestrial vertebrates (rat) were used to supplement the data set. A relevant population
level effect (reproduction) was observed in this species. Yet because the study used to develop the
hazard value was conducted in rats, which are less ecologically relevant than wildlife vertebrate species,
a moderate confidence was assigned to the relevance consideration for the terrestrial vertebrate
assessment.
No studies were reasonably available for the terrestrial invertebrate assessment of DIBP. Therefore, a
read-across was conducted from DBP (U.S. EPA. 2024a). Two studies from the analog DBP contained
endpoints that identified definitive hazard values below the DIBP limit of water solubility for two soil
invertebrate species. A moderate confidence was assigned to the quality of the database because two
terrestrial invertebrate species were represented by one high and one medium-rated study. In these two
Page 20 of 54
-------
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
PUBLIC RELEASE DRAFT
December 2024
species, the springtail (Folsomia fimetarid) and earthworm (Eisenia fetida), multiple endpoints were
identified. While no inconsistencies were observed in the data, the most sensitive endpoint that was used
to derive a hazard value was a 21-day EC 10 in the springtail and since no other studies contained
comparable endpoints, a moderate confidence evaluation was assigned to the consistency criterion. Due
to a clear dose-response relationship and a strong biologically relevant effect in the DBP data set for soil
invertebrates, a robust confidence was assigned to the strength and precision and dose-response criteria
for the soil invertebrate assessment.
No studies were available for the terrestrial plant assessment of DIBP. Therefore, a read-across was
conducted using DBP (U.S. EPA. 2024a). Most of the studies in the DBP database characterized doses
in a way that was not useful for developing a hazard value (e.g., in mg/m3 soil fumigation). Therefore,
slight confidence was assigned to the quality of the database. Since consistent growth effects were seen
in a variety of species, but the observed effects were distributed over a wide range of concentrations, a
moderate confidence was assigned to the consistency consideration. A dose-response effect was
observed in the study used to derive a hazard threshold, but a clear dose response was not observed in all
studies. Due to the added uncertainty from some studies in similar plants showing a lack of strong
biologically relevant effects or clear dose-response, moderate confidence was assigned to the strength
and precision and dose-response considerations for the terrestrial plants assessment.
Table 5-1. DIBP Evidence Table Summarizing the Overall Confidence Derived from Hazard
Thresholds
Types of Evidence
Quality
of the
Database
Consistency
Strength and
Precision
Biological
Gradient/Dose-
Response
Relevance
Hazard
Confidence
Aquatic
Acute Aquatic (SSD)
+++
+++
+++
+++
+++
Robust
Chronic Aquatic
Vertebrates
+++
++
+++
+++
++
Robust
Chronic Aquatic
Invertebrates
+++
+++
+++
+++
++
Robust
Chronic Benthic
Invertebrates
+
+++
+++
++
++
Moderate
Aquatic Plants & Algae
+
+++
++
++
++
Moderate
Terrestrial
Terrestrial Vertebrates
+++
++
+++
+++
++
Moderate
Terrestrial Invertebrates
++
+++
+++
++
Robust
Terrestrial Plants
+
++
f+
Moderate
3 Relevance includes biological, physica
/chemical, and environmenta
relevance.
+++ Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The
supporting weight of the scientific evidence outweighs the uncertainties to the point where it is unlikely that the
uncertainties could have a significant effect on the hazard estimate.
++ Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting
scientific evidence weighed against the uncertainties is reasonably adequate to characterize hazard estimates.
+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the
scenario, and when the assessor is making the best scientific assessment possible in the absence of complete
information. There are additional uncertainties that may need to be considered.
Page 21 of 54
-------
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
PUBLIC RELEASE DRAFT
December 2024
6 ENVIRONMENTAL HAZARD THRESHOLDS
EPA calculated hazard thresholds to identify potential concerns to aquatic and terrestrial species. After
weighing the scientific evidence, EPA selected the appropriate toxicity value from the integrated data to
use for hazard thresholds. Table 6-1 summarizes the aquatic concentrations of concern and Table 6-2
summarizes the terrestrial hazard values identified for DIBP. See Appendix C for more details about
how EPA weighed the scientific evidence.
Aquatic Organism Threshold
For aquatic species, EPA uses probabilistic approaches (e.g., SSD) when enough data are available
(eight or more species) and deterministic approaches (e.g., deriving a geometric mean of several
comparable values) when limited data are available. A SSD is a type of probability distribution of
toxicity values from multiple species. It can be used to visualize which species are most sensitive to a
toxic chemical exposure, and to predict a concentration of a toxic chemical that is hazardous to a
percentage of test species. This hazardous concentration is represented as an HCp, where p is the percent
of species below the threshold. EPA used an HC05 (a Hazardous Concentration threshold for 5 percent
of species) to estimate a concentration that would protect 95 percent of species. This HC05 can then be
used to derive a COC, and the lower bound of the 95 percent CI of the HC05 can be used to account for
uncertainty instead of dividing by an AF. For chronic exposures, an AF of 10 is used to account for
uncertainty associated with increased exposure duration. EPA has more confidence in the probabilistic
because an HC05 is representative of a larger portion of species in the environment. For the
deterministic approach, a COC is calculated by dividing a hazard value by an AF according to EPA
methods (U.S. EPA. 2016b. 2013. 2012V
Equation 6-1
COC = toxicity value -h AF
Terrestrial Organism Threshold
For terrestrial species, EPA estimates hazard by calculating a toxicity reference value (TRV), in the case
of terrestrial mammals and birds, or by assigning the hazard value as the hazard threshold in the case of
terrestrial plants and soil invertebrates. The TRVs generated for EPA's ecological soil screening levels
(Eco-SSLs) are defined as doses, "above which ecologically relevant effects might occur to wildlife
species following chronic dietary exposure and below which it is reasonably expected that such effects
will not occur" (U.S. EPA. 2007. 2005a). EPA prefers to derive the TRV by calculating the geometric
mean of the NOAELs across sensitive endpoints (growth and reproduction) rather than using a single
endpoint. The TRV method is preferred because the geometric mean of NOAELs across studies, species,
and endpoints provides greater representation of environmental hazard to terrestrial mammals and/or
birds. However, when the criteria for using the geometric mean of the NOAELs as the TRV are not met,
the TRVs for terrestrial mammals and birds are derived using a single endpoint. Due to a lack of
available terrestrial data for DIBP, EPA used a deterministic approach and assigned a hazard value
based on the most sensitive endpoint for each taxa.
6.1 Aquatic Species COCs
EPA derived three acute aquatic COCs and three chronic COCs using a combination of probabilistic and
deterministic approaches with DIBP hazard data supplemented with a read-across from DBP. Plant and
algae data was assessed separately and not incorporated into acute or chronic COCs because durations
normally considered acute for other species (e.g., up to 96 hours) can encompass several generations of
algae. Section 3 summarizes the aquatic hazard thresholds.
Page 22 of 54
-------
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
PUBLIC RELEASE DRAFT
December 2024
Acute Aquatic Organism Threshold
The aquatic acute COC for DIBP was derived from an SSD that contained 96-h LC50s for nine species
identified in systematic review (two species with DIBP hazard data and seven species with DBP hazard
data), bolstered by an additional 72 predicted LC50 values from the Web-ICE toxicity value estimation
tool. All studies included in the SSD were rated high or medium quality. After reviewing the possible
statistical distributions for the SSD, the Metropolis Hastings was chosen with a Logistic distribution.
This choice was based on an examination of p-values for goodness of fit, visual examination of Q-Q
plots, and evaluation of the line of best fit near the low-end of the SSD. The HC05 for this distribution is
406 |ig/L. After taking the lower 95th percentile of this HC05 as an alternative to the use of assessment
factors, the acute aquatic COC for vertebrates and invertebrates is 287 jug/L. See Appendix B for details
of the SSD that was used to derive the acute aquatic COC for DIBP. The SSD-derived acute aquatic
COC is similar to the multiomics-based PODs derived by EPA (Bencic et al.. 2024). Specifically, the
PODs derived by EPA ranged from 150 |ig /L (mPOD) to 900 |ig /L (bPOD) (Tble Apx D-l).
Chronic Aquatic Vertebrate Threshold
No chronic aquatic vertebrate studies were available for the quantitative assessment of potential hazards
from DIBP exposure. Therefore, analog data on chronic aquatic vertebrate hazards from DBP exposure
were used in a read-across to DIBP. The hazard value chosen to derive a hazard threshold resulted from
a high-quality rated study on the Japanese medaka (Oryzias latipes) (EAG Laboratories. 2018). In this
multi-generational study, the growth of F1 and F2 generations were significantly affected by exposure to
DBP. Specifically, there was significant inhibition of body weight at the lowest concentration studied in
the male F1 generation, with an unbounded LOEC value of 15.6 |ig/L DBP. In the female F1 generation,
the ChV for bodyweight inhibition was 0.082 mg/L DBP. In the F2 generation, the ChV for bodyweight
inhibition in male fish was 0.117 mg/L DBP and 24.6 |ig/L DBP in females. The most sensitive
endpoints in this data set were for inhibition of bodyweight in F1 males (0.0015 mg/L) and F2 females
(0.0246 mg/L). However, there was not a clear dose-response relationship for the body weight inhibition
response as some of the higher concentrations of DBP displayed a smaller mean effected compared to
the lower doses. Thus, this endpoint was not considered for the derivation of a COC. The most sensitive
endpoint for which there was a reliable dose-response relationship between DBP exposure and reduced
body weight was in F1 male fish, with a 112-day unbounded LOEC of 15.6 |ig/L DBP. At this
concentration, body weight was inhibited by 13.4 percent compared to the control and there was a clear
dose-response relationship up to the highest concentration tested of 304 |ig/L in which there was a 34
percent inhibition of body weight. Therefore, the hazard value was found to be 15.6 |ig/L and after
dividing by an AF of 10, the chronic aquatic vertebrate threshold is 1.56 jug/L.
Chronic Aquatic Invertebrate Threshold
No chronic aquatic invertebrate studies were available for the quantitative assessment of potential
hazards from DIBP exposure. Therefore, analog data on chronic aquatic invertebrate hazards from DBP
exposure were used in a read-across to DIBP. The most sensitive hazard value resulted from a medium-
quality rated study on the marine amphipod crustacean (Monocorophium acherusicum), which identified
a 14-day ChV of 0.122 mg/L DBP for reduced population abundance (Tagatz et al.. 1983). In this study,
crustacean abundance was reduced by 91 percent at 0.340 mg/L resulting in aNOEC/LOEC of
0.044/0.340 mg/L DBP. The 14-day ChV for reduction in population abundance in the marine amphipod
crustacean was selected to derive the chronic COC for aquatic invertebrates. After applying an AF of 10,
the chronic COC for aquatic invertebrates is 12.23 jug/L.
Acute Aquatic Benthic Invertebrate Threshold
The acute aquatic COC (287 ng/L) encompasses the level of concern for benthic invertebrates as it was
Page 23 of 54
-------
762
763
764
765
766
767
768
769
770
111
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
PUBLIC RELEASE DRAFT
December 2024
derived from an SSD that contained empirical data from the DIBP data set, read-across data from the
DBP data set, as well as Web-ICE-derived predicted LC50s for several benthic species including worms
(.Lumbricuius variegatus), snails (Physella gyrina, Lymnaea stagrialis), and copepods (Tigriopus
japonicus) (See Appendix B).
Chronic Aquatic Benthic Invertebrate Threshold
No chronic aquatic benthic invertebrate studies were available for the quantitative assessment of
potential hazards from DIBP exposure. Therefore, analog data on chronic aquatic benthic invertebrate
hazards from DBP exposure were used in a read-across to DIBP. The most sensitive hazard value
resulted from a high-quality rated study on the midge (Chironomus tentans) (Call et al.. 2001). In this
study, a 10-day ChV for population loss of 1,143.3 mg DBP/kg dry sediment in medium-TOC sediments
(4.80 percent) was identified. This study was conducted with low, medium, and high TOC sediments
and toxicity was found to decrease with an increase in sediment TOC. This endpoint for deriving the
COC using a medium-TOC was chosen because it is the closest to the assumed TOC level (4 percent)
used in Point Source Calculator (EPA. 2019) to estimate DBP exposure in benthic organisms. At the
LOEC identified in the study, 3,090 mg DBP/kg dry sediment, the midge population was reduced by
76.7 percent. Therefore, this endpoint was considered acceptable to derive a COC because of
population-level relevance and a clear dose-response relationship. After dividing by an AF of 10, the
chronic COC for benthic invertebrates is 114.3 mg/kg dry sediment.
Aquatic Algae Threshold
No aquatic plant and algae studies were available for the quantitative assessment of potential hazards
from DIBP exposure. Therefore, analog data on aquatic plant and algae hazards from DBP exposure
were used in a read-across to DIBP. The most sensitive endpoint resulted from a medium-quality green
algae {Selenastrum capricornutum) study (Adachi et al.. 2006) with DBP concentrations ranging from
0.1 to 10 mg/L. In this study, algal population was found to be reduced at 1.0 mg/L. Thus, a 96-hour
NOEC/LOEC of 0.1/1.0 mg/L, and a ChV of 0.316 mg/L was calculated. A clear dose-response
relationship was observed and therefore this endpoint was considered acceptable to derive a COC. After
dividing by an AF of 10, the COC for aquatic plants and algae is 31.6 jug/L.
Table 6-1. Aquatic Environmental Hazard Threshold for
DIBP
Receptor Group
Exposure
Scenario
Phthalate
Hazard
Threshold
(COC)
Citation
Aquatic
Vertebrates
Acute
DIBP and DBP
287 ng/L
SSD (See Section 3)
Chronic
DBP
1.56 ng/L
(EAG Laboratories,
2018)
Aquatic
Invertebrates
Acute
DIBP and DBP
287 |ig/L
SSD (See Section 3)
Chronic
DBP
12.23 |ig/L
(Tasatz et al., 1983)
Benthic
Invertebrates
Acute
DIBP and DBP
287 |ig/L
SSD (See Section 3)
Chronic
DBP
114.3 mg/kg dry
sediment
(Call et al.. 2001)
Aquatic Plants and
Algae
NA
DBP
31.6 ng/L
(Adachi et al., 2006)
Page 24 of 54
-------
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
PUBLIC RELEASE DRAFT
December 2024
6.2 Terrestrial Species Hazard Values
Terrestrial mammal threshold
EPA reviewed two laboratory rodent studies as surrogates for hazards of DIBP to wild mammal
populations (Saillenfait et al.. 2008; Saillenfait et al.. 2006). The most sensitive endpoint resulted from
one study in which pregnant Sprague-Dawley rates were given DIBP at doses of 0 (olive oil), 250, 500,
750, and 1000 mg/kg/day via gavage. In both male and female fetuses, body weight was significantly
lower (nine percent) at 500 mg/kg/day compared to controls, resulting in a gestational day 20
NOEC/LOEC of 250/500 mg/kg/day (Saillenfait et al.. 2006). The ChV and thus the terrestrial mammal
hazard threshold is 353 mg/kg/day.
Terrestrial Invertebrate Threshold
No acceptable terrestrial invertebrate studies were available for the quantitative assessment of potential
hazards from DIBP exposure. Therefore, analog data on terrestrial invertebrate hazards from DBP
exposure were used in a read-across to DIBP. The most sensitive endpoint was found for the springtail
(Folsomiafimetarici) with a 21-d EC 10 of 14 mg DBP/kg dry soil for reduced reproduction (Jensen et
al.. 2001). This study was rated high quality. At the lowest concentration tested, 100 mg DBP/kg dry
soil, reproduction was reduced by approximately 60% at the lowest concentration tested. This endpoint
was considered acceptable to derive a hazard value because of population-level relevance and a clear
dose-response relationship. Hazard values for soil invertebrates are calculated as the geometric mean of
ChV, EC20, and EC 10 values for apical endpoints such as mortality, reproduction, or growth..
Therefore, the hazard threshold for terrestrial invertebrates is 14 mg DBP/kg dry soil.
Terrestrial Plant Threshold
No terrestrial plant studies were available for the quantitative assessment of potential hazards from
DIBP exposure. Therefore, analog data on terrestrial plant hazards from DBP exposure were used in a
read-across to DIBP. The hazard value used to derive a hazard threshold resulted from a high-quality
rated study on bread wheat (Triticum aestivum) (Gao et al.. 2019). In this study, a LOEL for reduction in
leaf and root biomass in bread wheat seedlings at 10 mg/kg dry soil was observed. There was a clear
dose-response observed, with biomass reduction increasing as the dose of DBP increased. At the highest
dose (40 mg/kg), root and leaf biomass were reduced by 29.93 and 32.10 percent, respectively. Since the
most sensitive endpoint in this study was an unbounded LOAEL, the actual threshold dose may have
been lower than the lowest dose studied. However, no information was available in the study to adjust
the value to account for this uncertainty. The hazard threshold for terrestrial plants for DBP derived
from this study is 10 mg/kg dry soil.
Table 6-2. Terrestrial Environmental Hazard Threshold for DIBP
Receptor Group
Data Source
Hazard Threshold
Citation
Terrestrial Mammals
DIBP
353 mg/kg/day
(Saillenfait et al., 2006)
Terrestrial Invertebrates
DBP
14 mg DIBP/kg dry soil
(Tsens et al., 2013)
Terrestrial Plants
DBP
10 mg DBP/kg dry soil
(Gao et al., 2019)
7 ENVIRONMENTAL HAZARD ASSESSMENT CONCLUSIONS
EPA considered all reasonably available information identified through the systematic review process
under TSCA to characterize environmental hazard endpoints for DIBP. The following bullets summarize
the hazard values:
Page 25 of 54
-------
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
PUBLIC RELEASE DRAFT
December 2024
Aquatic species:
o DIBP had few reasonably available data to assess aquatic hazard,
o Analog data from DBP were used in a read-across to DIBP aquatic hazard,
o LC50 values from nine studies with exposures to DIBP and DBP in fish and aquatic
invertebrates were used alongside Web-ICE hazard estimates to develop an SSD. The lower
confidence interval of the HC05 was used as the COC and indicated that acute toxicity occurs at
287 |ig/L for DIBP.
o The chronic aquatic vertebrate hazard threshold was derived from a read-across from DBP in
which a three-generational reproductive study in Japanese medaka found significantly reduced
body weight in F1 male fish after a 112-day exposure to DBP. The COC based on this study
indicated that chronic toxicity in aquatic vertebrates occurs at 1.56 |ig/L.
o The chronic aquatic invertebrate hazard threshold was derived from a read-across from DBP in
which a 14-day exposure to DBP in the marine amphipod crustacean found a significant
reduction in population abundance. The COC based on this study indicated that chronic toxicity
in aquatic invertebrates occurs at 12.23 |ig/L.
o The chronic aquatic benthic invertebrate hazard threshold was derived from a read-across from
DBP in which a 10-day study on the midge identified a reduction in population at DBP
concentrations in medium TOC. The COC based on this study indicated that chronic toxicity in
chronic aquatic benthic invertebrates occurs at 114.3 mg/kg dry sediment,
o The aquatic plant and algae hazard threshold was derived from a read-across from DBP in which
a 96-hour exposure to DBP in the green algae Selenastrum capricornatam found a significant
reduction in population growth. The COC based on this study indicated that toxicity in aquatic
plants and algae occurs at 31.6 |ig/L.
Terrestrial Species:
o Terrestrial wildlife mammalian hazard data were not available for DIBP or the analog DBP,
therefore studies in laboratory rats were used to derive hazard values. Empirical DIBP toxicity
data for rats were used to estimate a hazard value for terrestrial mammals at 353 mg/kg-bw/day.
o The terrestrial invertebrate hazard threshold was derived from a read-across from DBP in which
21-day study in the springtail (Folsomia fimetaria) exposed to DBP via soil identified significant
effects on reproduction. The hazard threshold based on this study indicated that toxicity in
terrestrial invertebrates occurs at 14 mg DBP/kg dry soil,
o The terrestrial plant hazard threshold was derived from a read-across from DBP in which a
reduction in leaf and root biomass in bread wheat seedlings exposed to DBP via soil was
observed. The hazard threshold based on this study indicated that toxicity in terrestrial plant
occurs at 10 mg DBP/kg dry soil.
Page 26 of 54
-------
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
PUBLIC RELEASE DRAFT
December 2024
REFERENCES
Abt Associates. (2021). Final use report for di-isononyl phthalate (DINP) - (1,2-benzene-dicarboxylic
acid, 1,2-diisononyl ester, and 1,2-benzenedicarboxylic acid, di-C8-10-branched alkyl esters, C9-
rich) (CASRN 28553-12-0 and 68515-48-0). (EPA-HQ-OPPT-2018-0436-0035). Washington,
DC: U.S. Environmental Protection Agency. https://www.regulations.gov/document/EPA-HQ-
QPPT-2018-043 6-003 5
Adachi. A: Asa. K; Okano. T. (2006). Efficiency of rice bran for removal of di-n-butyl phthalate and its
effect on the growth inhibition of Selenastrum capricornutum by di-n-butyl phthalate. Bull
Environ Contam Toxicol 76: 877-882. http://dx.doi.org/10.1007/s00128-006-100Q-4
Adams. WJ: Biddinger. GR; Robillard. KA; Gorsuch. JW. (1995). A summary of the acute toxicity of 14
phthalate esters to representative aquatic organisms. Environ Toxicol Chem 14: 1569-1574.
http://dx.doi.org/10.1002/etc.562014Q916
ASTM. (2014). Standard guide for conducting acute toxicity tests on test materials with fishes,
macroinvertebrates, and amphibians. (E 29-96(2014)). West Conshohocken, PA.
https://compass.astm. org/document/?contentCode=ASTM%7CE0729-96R14%7Cen-US
Belford. B. (2023). 6.2: Similarity coefficients [Encyclopedia], In Cheminformatics. Davis, CA:
LibreTexts.
https://chem.libretexts.org/Courses/Intercollegiate Courses/Cheminformatics/06%3A Molecular
Similaritv/6.02%3A Similarity Coefficients
Bencic. DC: Flick. RW: Bell. ME: Henderson. WM; Huang. W: Purucker. ST: Glinski. DA: Blackwell.
BR: Christen. CH; Stacy. EH: Biales. AD. (2024). A multiomics study following acute exposures
to phthalates in larval fathead minnows (Pimephales promelas) - The potential application of
omics data in risk evaluations under TSCA (internal use only). (EPA/600/X-24/098). Cincinnati,
OH: U.S. Environmental Protection Agency.
Buccafusco. RJ: Ells. SJ: Leblanc. GA. (1981). Acute toxicity of priority pollutants to bluegill (Lepomis
macrochirus). Bull Environ Contam Toxicol 26: 446-452. http://dx.doi.org/10.1007/BFQ1622118
Burnham. KP; Anderson. DR. (2002). Model selection and multimodel inference: a practical
information-theoretic approach (2nd ed.). New York: Springer.
http://www.springer.com/statistics/statistical+theory+and+methods/book/978-0-387-95364-9
Call. DJ: Cox. DA: Geiger. PL: Genisot. KI; Markee. TP: Brooke. LT; Polkinghorne. CN: Vandeventer.
FA: Gorsuch. JW: Robillard. KA: Parkerton. TF: Reilev. MC: Anklev. GT: Mount. DR. (2001).
An assessment of the toxicity of phthalate esters to freshwater benthos. 2. Sediment exposures.
Environ Toxicol Chem 20: 1805-1815. http://dx.doi.org/10.1002/etc.562020Q826
Chen. X: Xu. S: Tan. T: Lee. ST: Cheng. SH: Lee. FWF: Xu. SJL: Ho. KC. (2014). Toxicity and
estrogenic endocrine disrupting activity of phthalates and their mixtures. Int J Environ Res
Public Health 11: 3156-3168. http://dx.doi.org/10.3390/iierphl 10303156
DeFoe. PL: Holcombe. GW: Hammermeister. DE: Biesinger. KE. (1990). Solubility and toxicity of
eight phthalate esters to four aquatic organisms. Environ Toxicol Chem 9: 623-636.
EAG Laboratories. (2018). Pibutyl phthalate: Medaka extended one generation reproduction test (final
report). (83260). Washington, PC: U.S. Environmental Protection Agency.
EG&G Bionomics. (1983a). Acute toxicity of fourteen phthalate esters to rainbow trout (Salmo
gairdneri) under flow-through conditions (final report) report no BW-83-3-1373 [TSCA
Submission], (Bionomics Report No. BW-83-3-1373. OTS0508403. 42005 B4-5. 40-8326144.
TSCATS/206776). Washington, PC: Chemical Manufacturers Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titlePetail/OTS05084Q3.xhtml
EG&G Bionomics. (1983b). Exhibit III: Acute toxicity of thirteen phthalate esters to bluegill (Lepomis
macrochirus) [TSCA Submission], In Exhibit III: Acute toxicity of thirteen phthalate esters to
fathead minnow (Pimephales promelas) under flow-through conditions. (Bionomics report No.
BW-83-3-1368. OTS0508481. 42005 G5-2. 40-8326129. TSCATS/038115). Washington, PC:
Page 27 of 54
-------
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
PUBLIC RELEASE DRAFT
December 2024
Chemical Manufacturers Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/QTS0508481.xhtml
EG&G Bionomics. (1984a). Acute toxicity of twelve phthalate esters to mysid shrimp (Mysidopsis
bahia) [TSCA Submission], (EPA/OTS Doc #40-8426078). Washington, DC: Chemical
Manufacturers Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS05084Q5.xhtml
EG&G Bionomics. (1984b). Acute toxicity of twelve phthalate esters to Paratanytarsus parthenogenica
(final report) report no BW-83-6-1424 [TSCA Submission], (EPA/OTS Doc #40-8426146).
Chemical Manufacturers Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS05084Q4.xhtml
EnviroSystem. (1991). Early life-stage toxicity of di-n-butyl phthalate (DnBP) to the rainbow trout
(Oncorhynchus mykiss) under flow-through conditions [TSCA Submission], (9102-CMA.
OTS0533141. 42005 L5-5. 40-9126399). Washington, DC: Chemical Manufacturers
Association. https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTSQ533141.xhtml
EPA. US. (2019). Point Source Calculator: A Model for Estimating Chemical Concentration in Water
Bodies. Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention.
Etterson. M. (2020). Species Sensitivity Distribution (SSD) Toolbox. Duluth, MN: U.S. Environmental
Protection Agency. Retrieved from https://www.epa.gov/sciencematters/species-sensitivitv-
distribution-toolbox-new-tool-identifv-and-protect-vulnerable
Gao. M; Guo. Z; Dong. Y; Song. Z. (2019). Effects of di-n-butyl phthalate on photosynthetic
performance and oxidative damage in different growth stages of wheat in cinnamon soils.
Environ Pollut 250: 357-365. http://dx.doi.org/10.1016/i.envpol.2019.04.022
Gu. S: Zheng. H; Xu. O: Sun. C: Shi. M; Wang. Z; Li. F. (2017). Comparative toxicity of the plasticizer
dibutyl phthalate to two freshwater algae. Aquat Toxicol 191: 122-130.
http://dx.doi.Org/10.1016/i.aquatox.2017.08.007
Jensen. J: van Langevelde. J: Pritzl. G: Krogh. PH. (2001). Effects of di(2-ethylhexyl) phthalate and
dibutyl phthalate on the collembolan Folsomia fimetaria. Environ Toxicol Chem 20: 1085-1091.
http://dx.doi.org/10.1002/etc.562020052Q
Jia. ZH; Yi. JH; Su. YR; Shen. H. (2011). Autotoxic substances in the root exudates from continuous
tobacco cropping. Allelopathy J 27: 87-96.
Lake Superior Research Institute. (1997). Sediment toxicity testing program for phthalate esters.
(Unpublished Report PE-88.0-SED-WIS). Arlington, VA: Chemical Manufacturers Association.
Lenoir. A: Touchard. A: Devers. S: Christides. JP; Boulav. R; Cuvillier-Hot. V. (2014). Ant cuticular
response to phthalate pollution. Environ Sci Pollut Res Int 21: 13446-13451.
http://dx.doi.org/10.1007/sll356-014-3272-2
Linden. E; Bengtsson. BE: Svanberg. O; Sundstrom. G. (1979). The acute toxicity of 78 chemicals and
pesticide formulations against two brackish water organisms, the bleak (Alburnus alburnus) and
the harpacticoid Nitocra spinipes. Chemosphere 8: 843-851. http://dx.doi.org/10.1016/0045-
653 5(79)90015-8
Liu. N: Wen. F; Li. F; Zheng. X: Liang. Z; Zheng. H. (2016). Inhibitory mechanism of phthalate esters
on Karenia brevis. Chemosphere 155: 498-508.
http ://dx. doi. org/10.1016/i. chemosphere.2016.04.082
McCarthy. JF; Whitmore. DK. (1985). Chronic toxicity of di-n-butyl and di-n-octyl phthalate to
daphnia-magna and the fathead minnow. Environ Toxicol Chem 4: 167-179.
http://dx.doi.org/10.1002/etc.56200402Q6
Mellor. CL; Marchese Robinson. RL; Benigni. R; Ebbrell. D; Enoch. SJ: Firman. JW; Madden. JC:
Pawar. G: Yang. C: Cronin. MTD. (2019). Molecular fingerprint-derived similarity measures for
toxicological read-across: Recommendations for optimal use. Regul Toxicol Pharmacol 101:
Page 28 of 54
-------
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
PUBLIC RELEASE DRAFT
December 2024
121-134. http://dx.doi.Org/10.1016/i.yrtph.2018.l 1.002
Pestana. CB; Firman. JW; Cronin. MTD. (2021). Incorporating lines of evidence from New Approach
Methodologies (NAMs) to reduce uncertainties in a category based read-across: A case study for
repeated dose toxicity. Regul Toxicol Pharmacol 120: 104855.
http://dx.doi.Org/10.1016/i.vrtph.2020.104855
Raimondo. S. .. D...N. Vivian. andM.G. Barron. (2010). Web-Based Interspecies Correlation Estimation
(Web-ICE) for Acute Toxicity: User Manual Version 3.1. (600R10004). Raimondo, S., D.N.
Vivian, and M.G. Barron. http://nepis.epa.gov/exe/ZyPURL.cgi?Dockev=P10068ND.txt
Rhodes. JE; Adams. WJ: Biddinger. GR; Robillard. KA; Gorsuch. JW. (1995). Chronic toxicity of 14
phthalate esters to Daphnia magna and rainbow trout (Oncorhynchus mykiss). Environ Toxicol
Chem 14: 1967-1976. http://dx.doi.org/10.1002/etc.5620141119
Saillenfait. AM: Sabate. JP; Gallissot. F. (2006). Developmental toxic effects of diisobutyl phthalate, the
methyl-branched analogue of di-n-butyl phthalate, administered by gavage to rats. Toxicol Lett
165: 39-46. http://dx.doi.Org/10.1016/i.toxlet.2006.01.013
Saillenfait. AM: Sabate. JP: Gallissot. F. (2008). Diisobutyl phthalate impairs the androgen-dependent
reproductive development of the male rat. Reprod Toxicol 26: 107-115.
http://dx.doi.Org/10.1016/i.reprotox.2008.07.006
Sevoum. A: Pradhan. A. (2019). Effect of phthalates on development, reproduction, fat metabolism and
lifespan in Daphnia magna. Sci Total Environ 654: 969-977.
http: //dx. doi. or g/10.1016/i. scitotenv .2018.11.158
Shen. C: Wei. J: Wang. T; Wang. Y. (2019). Acute toxicity and responses of antioxidant systems to
dibutyl phthalate in neonate and adult Daphnia magna. PeerJ 7: e6584.
http://dx.doi.org/10.7717/peeri.6584
Smithers Viscient. (2018). Di-n-butyl phthalate - short-term reproduction assay with fathead minnow
(Pimephales promelas) following OPPTS 890.1350 and OECD 229 guidelines. (Smithers
Viscient Study No. 13784.6123). Washington, DC: U.S. Environmental Protection Agency.
Springborn Bionomics. (1984). Chronic toxicity of fourteen phthalate esters to Daphnia magna with
cover letter dated 032585 [TSCA Submission] (pp. 95). (Report No. BW-84-5-1567.
OTS0000392-0. FYI-AX-0485-0392. TSCATS/032642). Wareham, MA: Chemical
Manufacturers Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS0000392Q.xhtml
Streufort. JM. (1978). Some effects of two phthalic acid esters on the life cycle of the midge
(Chironomus plumosus) [TSCA Submission], (OTS0000013-0. FYI-AX-1178-0013.
TSCATS/029296). Washington, DC: Manufacturing Chemists Association.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS0000013Q.xhtml
Tagatz. ME: Deans. CH: Moore. JC: Plaia. GR. (1983). Alterations in composition of field-developed
and laboratory-developed estuarine benthic communities exposed to di-normal-butyl phthalate.
Aquat Toxicol 3: 239-248. http ://dx. doi .org/10.1016/0166-445XC83 )90044-9
Tseng. IL; Yang. YF; Yu. CW: Li. WH: Liao. VHC. (2013). Phthalates induce neurotoxicity affecting
locomotor and thermotactic behaviors and AFD neurons through oxidative stress in
Caenorhabditis elegans. PLoS ONE 8: e82657. http://dx.doi.org/10.1371/iournal.pone.0082657
U.S. EPA. (1998). Guidelines for ecological risk assessment [EPA Report], (EPA/630/R-95/002F).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/risk/guidelines-ecological-risk-assessment
U.S. EPA. (2005a). Guidance for developing ecological soil screening levels [EPA Report], (OSWER
Directive 92857-55). Washington, DC: U.S. Environmental Protection Agency, Office of Solid
Waste and Emergency Response, http://www.epa.gov/chemical-research/guidance-developing-
ecological-soil-screening-levels
U.S. EPA. (2005b). Guidelines for carcinogen risk assessment [EPA Report], (EPA630P03001F).
Page 29 of 54
-------
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
PUBLIC RELEASE DRAFT
December 2024
Washington, DC. https://www.epa.gov/sites/production/files/2013-
09/documents/cancer guidelines final 3-25-05.pdf
U.S. EPA. (2007). Attachment 4-3 Guidance for Developing Ecological Soil Screening Levels (Eco-
SSLs) Eco-SSL Standard Operating Procedure (SOP) #4: Wildlife Toxicity Reference Value
Literature Review, Data Extraction and Coding. (OSWER9285755F).
http://nepis. epa.gov/exe/ZyPURL. cgi?Dockev=P 100CDHC.txt
U.S. EPA. (2012). Sustainable futures: P2 framework manual [EPA Report], (EPA/748/B-12/001).
Washington DC. http://www.epa.gov/sustainable-futures/sustainable-futures-p2-framework-
manual
U.S. EPA. (2013). Interpretive assistance document for assessment of discrete organic chemicals.
Sustainable futures summary assessment [EPA Report], Washington, DC.
http://www.epa.gov/sites/production/files/2015-05/documents/05-iad discretes iune2013.pdf
U.S. EPA. (2016a). Ecological effects test guidelines: OCSPP 850.1075: Freshwater and saltwater fish
acute toxicity test [EPA Report], (EPA 712-C-16-007; EPA-HQ-OPPT-2009-0154-0035).
Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention. https://www.regulations.gov/document/EPA-HQ-OPPT-2009-0154-0035
U.S. EPA. (2016b). Weight of evidence in ecological assessment [EPA Report], (EPA/100/R-16/001).
Washington, DC: Office of the Science Advisor.
https://nepis.epa. gov/Exe/ZvPURL.cgi?Dockev=P100SFXR.txt
U.S. EPA. (2020a). Final scope of the risk evaluation for di-ethylhexyl phthalate (1,2-
benzenedicarboxylic acid, l,2-bis(2-ethylhexyl) ester); CASRN 117-81-7 [EPA Report], (EPA-
740-R-20-017). Washington, DC: Office of Chemical Safety and Pollution Prevention.
https://www.epa.gov/sites/default/files/2020-09/documents/casrn 117-81-7 di-
ethylhexyl phthalate final scope.pdf
U.S. EPA. (2020b). Final scope of the risk evaluation for di-isobutyl phthalate (1,2-benzenedicarboxylic
acid, l,2-bis(2-methylpropyl) ester); CASRN 84-69-5 [EPA Report], (EPA-740-R-20-018).
Washington, DC: Office of Chemical Safety and Pollution Prevention.
https://www.epa.gov/sites/default/files/2020-09/documents/casrn 84-69-5 di-
isobutyl phthalate final scope.pdf
U.S. EPA. (2020c). Final scope of the risk evaluation for dibutyl phthalate (1,2-benzenedicarboxylic
acid, 1,2-dibutyl ester); CASRN 84-74-2 [EPA Report], (EPA-740-R-20-016). Washington, DC:
Office of Chemical Safety and Pollution Prevention.
https://www.epa.gov/sites/default/files/2020-09/documents/casrn 84-74-
2 dibutyl phthalate final scope O.pdf
U.S. EPA. (2021a). Draft systematic review protocol supporting TSCA risk evaluations for chemical
substances, Version 1.0: A generic TSCA systematic review protocol with chemical-specific
methodologies. (EPA Document #EPA-D-20-031). Washington, DC: Office of Chemical Safety
and Pollution Prevention. https://www.regulations.gov/document/EPA-HQ-OPPT-2021-0414-
0005
U.S. EPA. (2021b). Final scope of the risk evaluation for di-isodecyl phthalate (DIDP) (1,2-
benzenedicarboxylic acid, 1,2-diisodecyl ester and 1,2-benzenedicarboxylic acid, di-C9-ll-
branched alkyl esters, ClO-rich); CASRN 26761-40-0 and 68515-49-1 [EPA Report], (EPA-740-
R-21-001). Washington, DC: Office of Chemical Safety and Pollution Prevention.
https://www.epa.gOv/svstem/files/documents/2021-08/casrn-26761-40-0-di-isodecvl-phthalate-
final-scope.pdf
U.S. EPA. (2024a). Draft Environmental Hazard Assessment for Dibutyl Phthalate (DBP). Washington,
DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2024b). Draft Environmental Release and Occupational Exposure Assessment for Di-
isobutyl Phthalate (DIBP). Washington, DC: Office of Pollution Prevention and Toxics.
Page 30 of 54
-------
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
PUBLIC RELEASE DRAFT
December 2024
U.S. EPA. (2024c). Draft physical chemistry and fate and transport assessment for Dibutyl Phthalate
(DBP). Washington, DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2024d). Draft physical chemistry and fate and transport assessment for Diisobutyl phthalate
(DIBP). Washington, DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2024e). Draft Physical Chemistry Assessment for Diisobutyl phthalate (DIBP). Washington,
DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2024f). Draft Systematic Review Protocol for Diisobutyl Phthalate (DIBP). Washington,
DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2024g). Environmental Hazard Assessment for Diisodecyl Phthalate (DIDP). Washington,
DC: Office of Pollution Prevention and Toxics. https://www.regulations.gov/document/EPA-
HO-QPPT-2024-0073
U.S. EPA. (2024h). Web-based Interspecies Correlation Estimation (Web-ICE) for Acute Toxicity: User
Manual, Version 4.0. (EPA/600/B-24/158). Gulf Breeze, FL: Office of Research and
Development, Gulf Ecosystem Measurement and Modeling Division.
https://www3.epa.gov/webice/iceManual.html
U.S. EPA. (2025a). Draft Non-cancer Human Health Hazard Assessment for Diisobutyl phthalate
(DIBP). Washington, DC: Office of Pollution Prevention and Toxics.
U.S. EPA. (2025b). Draft Risk Evaluation for Diisobutyl Phthalate (DIBP). Washington, DC: Office of
Pollution Prevention and Toxics.
Wei. J: Shen. O: Ban. Y; Wang. Y; Shen. C: Wang. T; Zhao. W: Xie. X. (2018). Characterization of
Acute and Chronic Toxicity of DBP to Daphnia magna. Bull Environ Contam Toxicol 101: 214-
221. http://dx.doi.org/10.1007/sQ0128-018-2391-8
Willming. MM: Lilavois. CR; Barron. MG: Raimondo. S. (2016). Acute toxicity prediction to
threatened and endangered species using Interspecies Correlation Estimation (ICE) models.
Environ Sci Technol 50: 10700-10707. http://dx.doi.org/10.1021/acs.est.6b03009
Page 31 of 54
-------
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
PUBLIC RELEASE DRAFT
December 2024
Appendix A Analog Selection for Environmental Hazard
DIBP environmental hazard data were only reasonably available for aquatic and benthic species exposed
under acute durations with use of laboratory mammalian hazard data as surrogate for terrestrial
mammalian wildlife hazard from DIBP exposure. No algal, chronic aquatic, chronic benthic, terrestrial
plant, soil invertebrate, or avian hazard data were identified for DIBP. Additionally, the acute aquatic
and acute benthic hazard data set for DIBP were limited to a single 24-hour water exposure in fathead
minnows and a 96-hour water exposure in copepod Nitocra spinipes. Therefore, analog selection was
performed to identify an appropriate analog to read across to DIBP to supplement the aquatic, benthic,
terrestrial plant, soil invertebrate, and avian hazard data. Dibutyl phthalate (DBP) was selected as an
analog for read-across of aquatic, benthic, and soil invertebrate hazard data based on excellent structural
similarity, similar physical, chemical, environmental fate and transport behavior in water and sediment,
and similar ecotoxicological behavior in aquatic taxa, including mechanistic hazard comparisons in the
form of transcriptomic and metabolomic points of departure (Figure Apx A-l). DBP was also selected
for read-across of terrestrial plant and avian hazard, however, confidence in DBP as an analog for DIBP
was decreased for read-across to these two taxa. This is because terrestrial plant and avian
ecotoxicological similarity between DBP and DIBP could not be determined using the same means as in
the aquatic, benthic, and soil invertebrate hazard analog selection, therefore the terrestrial plant and
avian hazard read-across from DBP to DIBP was reliant upon similarity in structure as well as physical,
chemical, environmental fate and transport. The DBP environmental hazard data to be used as analog
data for DIBP received overall quality determinations of high or medium and are described in Section 3.
The similarities between DIBP and analog DBP are described in detail below.
Page 32 of 54
-------
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
PUBLIC RELEASE DRAFT
December 2024
FigureApx A-l. Framework for DIBP Environmental Hazard Analog Selection.
*Criterion may be relaxed to results in fewer programs if no analogs are generated by one or more
programs. AECOSAR acute and chronic toxicity predictions for vertebrates and invertebrates generated
for chemicals with log Kow < 5 and chronic toxicity predictions generated if log Kow < 8, and algal
toxicity predictions generated if log Kow < 6.4 should the chemical meet the definition of an ECOSAR
class. **Weight of scientific evidence and professional judgement involved in finalizing selection.
A.l Structural Similarity
Structural similarity between DIBP and candidate analogs was assessed using two TSCA New Approach
Methodologies (NAMs) (the Analog Identification Methodology (AIM) program and the Organisation
of Economic Cooperative Development Quantitative Structure Activity Relationship [OECD QSAR]
Toolbox) as well as two EPA Office of Research and Development tools (Generalized Read-Across
[GenRA]) and the Search Module within the Cheminformatics Modules). These four programs provide
Page 33 of 54
-------
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
PUBLIC RELEASE DRAFT
December 2024
complementary methods of assessing structural similarity. There are several different methods for
determining structural similarity. A fragment-based approach (e.g., as implemented by AIM) searches
for compounds with similar structural moieties or functional groups. EPA's TSCA New Chemicals
Program utilizes CBI-AIM to identify analogs with data (including analogs with CBI). CBI information
is not found in the public-facing version of AIM in order to protect business confidentiality, and CBI-
AIM has undergone updates not found in the public-facing version of AIM. A structural identifier
approach (e.g., the Tanimoto coefficient) calculates a similarity coefficient based on molecular
fingerprinting (Belford. 2023). Molecular fingerprinting approaches look at similarity in atomic pathway
radius between the analog and target chemical substance (e.g., Morgan fingerprint in GenRA which
calculates a Jaccard similarity index). Some fingerprints may be better suited for certain characteristics
and chemical classes. For example, substructure fingerprints like PubChem fingerprints perform best for
small molecules such as drugs, while atom-pair fingerprints, which assigns values for each atom within
a molecule and thus computes atom pairs based on these values, are preferable for large molecules.
Some tools implement multiple methods for determining similarity. Regarding programs which generate
indices, it has been noted that because the similarity value is dependent on the method applied, that these
values should form a line of evidence rather than be utilized definitively (Pestana et al.. 2021; Mellor et
al.. 2019).
AIM analogs were obtained using the Confidential Business Information (CBI) version of AIM and
described as 1st or 2nd pass (only analogs not considered CBI are included in TableApx A-2).
Tanimoto-based PubChem fingerprints were obtained in the OECD QSAR Toolbox (v4.4.1, 2020) using
the Structure Similarity option and are presented as a range. Chemical Morgan Fingerprint scores were
obtained in GenRA (v3.1) (limit of 100 analogs, no ToxRef filter). Tanimoto scores were obtained in the
Cheminformatics Search Module using Similar analysis. AIM 1st and 2nd pass analogs were compiled
with the top 100 analogs with indices greater than 0.5 generated from the OECD QSAR Toolbox and the
Cheminformatics Search Module and indices greater than 0.1 generated from GenRA. These filtering
criteria are displayed in Table Apx A-l. Analogs that appeared in three out of four programs were
identified as potential analog candidates (Figure Apx A-l). Using these parameters, 25 analogs were
identified as potentially suitable analog candidates for DIBP based on structural similarity (Table Apx
A-2). The results for structural comparison of DIBP to DBP (CASRN 84-74-2), diethylhexyl phthalate
(DEHP, CASRN 117-81-7), diisodecyl phthalate (DIDP, CASRN 26761-40-0), and diisononyl phthalate
(DINP, CASRN 28553-12-0) are further described below due to those analog candidates having
completed data evaluation and extraction.
Table Apx A-l. Structure Program Filtering Criteria
Program
Index
Filtering parameters
Analog Identification
Methodology (AIM)
Fragment-based
1st or 2nd pass
OECD QSAR Toolbox
Tanimoto-based PubChem
fingerprints
Top 100 analogs > 0.5
Cheminformatics Search
Module
Similarity-type: Tanimoto
Top 100 analogs with index >
0.5
GenRA
Morgan Fingerprints
Top 100 analogs with index >
0.1 (ToxRef data filter off)
Page 34 of 54
-------
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
PUBLIC RELEASE DRAFT
December 2024
DBP, DEHP, DINP, and DIDP were indicated as structurally similar to DIBP in AIM (analogs were 1st
or 2nd pass), OECD QSAR Toolbox (PubChem features = 0.9-1), and the Cheminformatics Search
Module (Tanimoto coefficient = 0.84-0.90) (TableApx A-2). Additionally, DBP and DEHP were
indicated as structurally similar to DIBP in GenRA (Morgan Fingerprint = 0.48 and 0.51, respectively)
(Table Apx A-2). DBP was ultimately selected for read-across of aquatic, benthic, and terrestrial hazard
to DIBP based on the additional lines of evidence (physical, chemical, and environmental fate and
transport similarity and ecotoxicological similarity).
Table Apx A-2. Structural Similarity between DIBP and Analog Candidates which met Filtering
Chemical
CASRN
AIM
OECD
QSAR
Toolbox
Cheminformatics
GenRA
Count
DIBP (target)
84-69-5
Exact
Match
1.00
1.00
1.00
4
Di(2-ethylhexyl)
phthalate (DEHP)"
117-81-
7
1 st pass
0.90-1.00
0.90
0.51
4
Bis(2-propylheptyl)
phthalate
53306-
54-0
1 st pass
0.90-1.00
0.90
0.48
4
Butyl 2-ethylhexyl
phthalate
85-69-8
1 st pass
0.90-1.00
0.90
-
3
Isoamyl phthalate
605-50-
5
1 st pass
0.90-1.00
0.90
0.58
4
Diisodecyl phthalate
(DIDP)"
26761-
40-0
1 st pass
0.90-1.00
0.84
-
3
Diisooctyl phthalate
27554-
26-3
1 st pass
0.90-1.00
0.84
-
3
Diisononyl phthalate
(DINP)"
28553-
12-0
1 st pass
0.90-1.00
0.84
-
3
Di(2-ethyl-4-
methylpentyl)
phthalate
2229-
55-2
1 st pass
0.90
0.60
3
Di -n-propy lphthal ate
131-16-
8
2nd
pass
0.90-1.00
0.95
0.51
4
Dibutyl 1,2-
benzenedicarboxylate
(DBP)"
84-74-2
2nd
pass
0.90-1.00
0.90
0.48
4
Diethyl phthalate
84-66-2
2nd
pass
0.90-1.00
0.89
0.56
4
Page 35 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Chemical
CASRN
AIM
OECD
QSAR
Toolbox
Cheminformatics
GenRA
Count
Dipentyl phthalate
131-18-
0
2nd
pass
0.90-1.00
0.88
3
Dihexyl phthalate
84-75-3
2nd
pass
0.90-1.00
0.88
3
Di-n-octyl phthalate
117-84-
0
2nd
pass
0.90-1.00
0.88
3
Ditridecyl phthalate
119-06-
2
2nd
pass
0.90-1.00
0.88
-
3
Didodecyl phthalate
2432-
90-8
2nd
pass
0.90-1.00
0.88
-
3
Diundecyl phthalate
3648-
20-2
2nd
pass
0.90-1.00
0.88
-
3
Diheptyl phthalate
3648-
21-3
2nd
pass
0.90-1.00
0.88
-
3
Dinonyl phthalate
84-76-4
2nd
pass
0.90-1.00
0.88
-
3
Didecyl phthalate
84-77-5
2nd
pass
0.90-1.00
0.88
-
3
Dimethyl phthalate
131-11-
3
2nd
pass
0.90-1.00
0.84
-
3
Isobutyl benzoate
120-50-
3
2nd
pass
-
0.92
0.51
3
Terephthalic acid,
diisobutyl ester
18699-
48-4
2nd
pass
-
0.92
0.50
3
Di(2-methoxy ethyl)
phthalate
117-82-
8
-
0.90-1.00
0.87
0.49
3
Cyclohexyl 2-isobutyl
phthalate
5334-
09-8
-
0.90-1.00
0.84
0.61
3
a Analogs which have completed data evaluation and extraction are bolded.
1182
1183 A.2 Physical, Chemical, and Environmental Fate and Transport Similarity
1184 DIBP analog candidates from the structural similarity analysis were preliminarily screened based on
1185 similarity in log octanol-water partition coefficient (log Kow) obtained using EPI Suite (Figure Apx
Page 36 of 54
-------
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
PUBLIC RELEASE DRAFT
December 2024
A-1). For this screening step, DIBP, DBP, DEHP, DIDP, and DINP values were obtained from their
respective final scope documents (Abt Associates. 2021; U.S. EPA. 2021b. 2020a. b, c). Analog
candidates with log Kow values within one log unit relative to DIBP were considered potentially suitable
analog candidates for DIBP. This preliminary screening analysis narrowed the analog candidate list from
25 candidate analogs to 3 candidate analogs (TableApx A-3). One of the three candidate analogs was
DBP. A more expansive analysis of physical, chemical, environmental fate and transport similarities
between DIBP and DBP was conducted because DBP's hazard data had completed data evaluation and
extraction (Figure Apx A-l).
Table Apx A-3. Analog Candidates with Similar log Kow values to that of DIBP
Chemical
CASRN
Log Kow
DIBP (target)
84-69-5
4.34
DBP
84-74-2
4.53
Diisobutyl terephthalate
18699-48-4
4.46°
Cyclohexyl 2-isobutyl phthalate
5334-09-8
5.33°
a Values predicted using EPI Suite
b Analogs which have completed data evaluation and extraction are
bolded.
Physical, chemical, and environmental fate and transport similarities between DIBP and DBP were
assessed based on properties relevant to the to the aquatic, benthic, and soil compartments and are
shown in Table Apx A-4. Physical, chemical, and environmental fate and transport values for DIBP and
DBP are specified in the Draft Fate & Physical Chemistry Assessment for Diisobutyl Phthalate (DIBP)
(U.S. EPA. 2024e). Draft Fate Assessment for Diisobutyl Phthlate (DIBP) (U.S. EPA. 2024d). Draft
Fate & Physical Chemistry Assessment for Dibutyl Phthalate (DBP) (U.S. EPA. 2024c). DIBP and DBP
water solubilities are similar in value (6.2 mg/L and 11.2 mg/L, respectively) indicating both target and
analog are fairly insoluble in water. The selected octanol-water partition coefficients (log Kow) are very
similar in value (4.34 and 4.5 for DIBP and DBP, respectively), indicating relatively low affinity for
water and higher sorption potential to soils and sediments for target and analog. Degradation of DIBP
and DBP in both water and sediment is also similar, with almost complete aerobic biodegradation in
water within 4 weeks and slower anaerobic degradation in sediment (Table Apx A-4). Both DIBP and
DBP would biodegrade in water before hydrolyzing. Similar biodegradation rates between target and
analog can increase confidence when considering read across of chronic hazard. The values for DIBP's
and DBP's log organic carbon-water partition coefficients indicate both target and analog will be
preferentially bound to sediment or soil than exist in the water. Bioaccumulation potential of DBP in
aquatic organisms is slightly higher than for DIBP by one to two orders of magnitude (Table Apx A-4),
however both phthalates have fairly low bioconcentration and bioaccumulation potential. An almost
identical freshwater magnification factor of less than 1 was derived across 18 species for DIBP and DBP
indicating that both phthalates do not biomagnify up the trophic levels (Table Apx A-4). Regarding fate
in terrestrial species, bioconcentration of DIBP and DBP in various terrestrial plants is low (0.13-2.23
and 0.02-9.32, respectively, Table_Apx A-4). Almost identical uptake behavior was noted in ants
covered with either 2,000 ng DIBP or DBP (Lenoir et al.. 2014). DIBP's and DBP's vapor pressures are
very low (4.76x 10"5 mmHg and 2.01 x 10"5 mmHg, respectively) as are their Henry's law constants
Page 37 of 54
-------
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
PUBLIC RELEASE DRAFT
December 2024
(1.83xl0"7 atm-m3/mol and 1.81xl0"6 atm-m3/mol, respectively), indicating both chemicals are not
readily volatile. Both phthalates exist as a liquid at room temperature and have similar molecular
weights. The similarity in the properties described in TableApx A-4 support the ability to read across
DBP aquatic and benthic hazard as well as terrestrial (plant, soil invertebrate, and avian) hazard to
supplement the DIBP environmental hazard data set.
Table Apx A-4. Comparison of DIBP and DBP for Several Physical and Chemical and
Environmental Fate Properties I
televant to Water, Sediment, and Soil
Property
DIBP (target)
DBP
Water Solubility
6.2 mg/L
11.2 mg/L
Log Kow
4.34
4.5
Log Koc
2.67 (2.50-2.86)
3.69 (3.14-3.94)
Hydrolysis (ti 2)
5.3 yr (pH 7); 195 days (pH 8)
3.43 yr (pH 7); 125 days (pH 8)
Aerobic biodegradation in water
42 to 98% in 28 days
68.3 to >99% after 28 days
Anaerobic biodegradation in
sediment
0 - 30% after 56 days
ti/2 = 14.4 days
BCF
30.2 L/kg wet weight
(estimated)
2.9 - 176 (experimental, various
aquatic species)
BAF (aquatic)
30.2 L/kg wet weight
(estimated)
100 - 1,259 (experimental,
various fish species), 159
(estimated)
FWMF (aquatic)
0.81 (18 marine species)
0.70 (18 marine species)
BCF (plants)
0.13-2.23 (onion, celery,
pepper, tomato, bitter gourd,
eggplant, and long podded
cowpea)
0.02-9.32 (rice, radish, wheat,
maize, strawberry, carrot,
lettuce, wetland grasses)
Henry's Law Constant (atm-
m3/mol)
1.83xl0~7
1.81xl0~6
Vapor Pressure (mmHg)
4.76xl0~5
2.01xl0~5
Molecular Weight
278.35 g/mol
278.35 g/mol
Physical state of the chemical
Clear Viscous Liquid
Clear Oily Liquid
A.3 Ecotoxicological Similarity
Ecotoxicological similarity between DIBP and DBP was assessed based on two lines of evidence: the
first line of evidence was a comparison of the analog's empirical hazard data to corresponding toxicity
predictions of the target and the second line of evidence was a comparison of several points of departure
derived for DIBP and DBP following acute exposures to fathead minnow and copepod Nitocra spinipes
(Bencic et al.. 2024; Linden et al.. 1979). Although less relevant than hazard obtained from sediment
exposures, toxicological similarity in empirical hazard evidence for aquatic invertebrates exposed to
Page 38 of 54
-------
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
PUBLIC RELEASE DRAFT
December 2024
DIBP and DBP in water was also assessed to determine suitability of DBP for read-across of soil
invertebrate hazard data to DIBP. Ecotoxicological comparisons made for algae helped support the read-
across for terrestrial plant hazard, while acknowledging the differences between nonvascular aquatic
biota and vascular terrestrial plants. DIBP toxicity predictions for acute and chronic exposure to fish,
aquatic invertebrates, and green algae were generated using ECOSAR v2.2. Empirical hazard data used
in the following comparisons were from studies with overall quality determinations of high and medium.
The ecotoxicological similarity line of evidence had uncertainty in supporting the avian hazard read-
across from DBP to DIBP due to a lack of predictive tools for assessing this hazard, therefore less
confidence was had in the avian read-across.
Comparison of the analog empirical hazard data to corresponding ECOSAR toxicity predictions for
DIBP shows agreement of hazard values well within 10-fold (Figure Apx A-l, TableApx A-5).
Average ratio of empirical DBP aquatic hazard data to predicted DIBP hazard values is 1.3 ± 0.20
(standard error) (Table Apx A-5) which indicates very similar ecotoxicological behavior between DBP
and DIBP when aquatic vertebrates, aquatic invertebrates, and algae are exposed under acute and
chronic conditions and that DBP is an appropriate analog for DIBP. An additional comparison based on
DIBP and DBP empirical hazard from the same studies also indicate ecotoxicological similarity between
DIBP and DBP. Transcriptomic, metabolomic, and swimming behavior points of departure as well as
LC50 values were derived for DIBP and DBP following a 24-hour exposure to fathead minnow (Bencic
et al.. 2024). In a second study, 96-hour LC50 values were derived for benthic invertebrate N. spinipes
exposed to DIBP and DBP (Linden et al.. 1979). Although the DIBP and DBP hazard values are within
10-fold of each other and suggest general agreement, the lower hazard values for DBP when compared
to DIBP indicate the analog data is protective of the target when both phthalates are tested in the same
study across two aquatic taxa, with average ratio of DBP hazard to DIBP hazard 0.38 ± 0.11 (standard
error) (Table Apx A-6). These comparisons support the appropriateness to read-across DBP aquatic and
benthic hazard data to DIBP. Ecotoxicological similarity for a soil invertebrate hazard read-across is
inferred by the aquatic and benthic invertebrate toxicity comparisons made between DIBP and DBP,
similar to the read-across approach used for other phthalates (U.S. EPA. 2024g).
Table Apx A-5. Ecotoxicological similarity in aquatic taxa exposed to DIBP (predicted hazard)
and DBP (empirical hazard)
Taxa
Duration
Endpoint
DIBP
DBP
Ratio of DBP
toxicity to
DIBP toxicity
Predicted
hazard (mg/L)°
Empirical hazard
(mg/L)°
Fish
96-h
LC50
1.30
1.086
0.8
Daphnid
48-h
LC50
2.57
3.44c
1.3
Mysid
96-h
LC50
0.98
0.61c/
0.6
Green Algae
96-h
EC50
0.82
1.12e
1.4
Fish
ChV
0.11
0.1 of
0.9
Daphnid
ChV
0.54
1.14g
2.1
Green Algae
ChV
0.31
0.56/1
1.8
Page 39 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Taxa
Duration
Endpoint
DIBP
Predicted
hazard (mg/L)"
DBP
Empirical hazard
(mg/L)°
Ratio of DBP
toxicity to
DIBP toxicity
Average fold-hazard DBPrDIBP
1.3 ± 0.20
° Hazard values, including empirical hazard values used to calculate a geometric mean, were limited
to those at or below the phthlate-specific water solubility limit.
b Value for DBP represents a geometric mean of 96-hour fish {Lepomis macrochirus, Pimephales
pr omelets, Oncorhynchus my kiss) LC50 data from (Smithers Viscient. 2018; Adams et al.. 1995;
DeFoe et al.. 1990; McCarthy and Whitmore. 1985; EG&G Bionomics. 1983b; Buccafusco et al..
1981).
c Value for DBP represents a geometric mean of 48-hour Daphnia magna LC50 and 48-hour Daphnia
magna immobilization EC50 data from (Shen et al.. 2019; Wei et al.. 2018; Adams et al.. 1995;
McCarthy and Whitmore. 1985).
d Value for DBP represents a geometric mean of 96-hour Americamysis bahia LC50 data from
(Smithers Viscient. 2018; Adams et al.. 1995; DeFoe et al.. 1990; McCarthy and Whitmore. 1985;
EG&G Bionomics. 1983b; Buccafusco etal.. 1981).
c Value for DBP represents a geometric mean of 96-hour green algae (Selenastrum capricornutum and
Chlorellapyrenoidosa) EC50 data from (Gu et al.. 2017; Adams et al.. 1995).
' Value for DBP represents a geometric mean of fish {Lepomis macrochirus, Pimephalespromelas,
Oncorhynchus mykiss, Oryzias latipes) NOEC/LOEC pairs for Mortality, Reproduction, and
Development/Growth endpoints from (Smithers Viscient. 2018; Adams et al.. 1995; DeFoe et al..
1990; McCarthy and Whitmore. 1985; EG&G Bionomics. 1983b; Buccafusco et al.. 1981). Exposures
and study durations were a minimum of 13 days.
g Value for DBP represents a geometric mean of Daphnia magna NOEC/LOEC pairs for Mortality,
Reproduction, and Development/Growth endpoints from (Sevoum and Pradhan. 2019; Wei et al..
2018; Rhodes et al.. 1995; DeFoe et al.. 1990; Springborn Bionomics. 1984). Exposures and study
durations were a minimum of 1 week and 2 weeks, respectively.
h Value for DBP represents a geometric mean of green algae {Selenastrum capricornutum)
NOEC/LOEC pairs for Development/Growth endpoints from (Adachi et al.. 2006).
Table Apx A-6. Comparison of DIBP and DBP Points of Departure and LC50 Values in Fathead
Species
Outcome
Endpoint
DIBP
Hazard
(mg/L)
DBP
Hazard
(mg/L)
Ratio of DBP
toxicity to
DIBP toxicity
Fathead minnow0
Transcriptomics
POD
0.87
0.12
0.14
Fathead minnow0
Metabolomics
POD
0.15
0.11
0.73
Fathead minnow0
Swimming
behavior
POD
0.90
0.24
0.27
Fathead minnow0
Mortality
LC50
5.30
1.02
0.19
Nitocra spinipesb
Mortality
LC50
3.0
1.7
0.57
Page 40 of 54
-------
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
PUBLIC RELEASE DRAFT
December 2024
Species
Outcome
Endpoint
DIBP
Hazard
(mg/L)
DBP
Hazard
(mg/L)
Ratio of DBP
toxicity to
DIBP toxicity
Average fold-hazard DBPrDIBP
0.38 ± 0.11
a Data are based on measured concentrations from (Bencic et al.. 2024).
b Data are based on measured concentrations from (Linden et al.. 1979).
A.4 Read-Across Weight of the Scientific Evidence and Conclusions
DIBP presented with minimal acute aquatic and benthic hazard data, no chronic aquatic or chronic
benthic hazard data, no algal hazard data, and no terrestrial plant, soil invertebrate, or avian hazard data.
Analog selection was carried out to address these data gaps. Several phthalates of interest (DBP, DEHP,
DIDP, and DINP) were indicated as structurally similar to DIBP. A screening by logKow values and
further comparison of additional physical, chemical, and environmental fate and transport properties
indicated that DBP, which is data-rich for aquatic and benthic hazard, was very similar to DIBP. A
comparison of available DBP empirical hazard data to corresponding DIBP toxicity predictions for
aquatic taxa showed high concordance between analog and target hazard. A second toxicity comparison
was made in fathead minnow and copepod N. spinipes exposed to either DBP or DIBP for 24 hours
(fathead minnow) or 96 hours (N. spinipes); DBP points of departure and LC50 in fathead minnow were
within 10-fold of and protective of DIBP points of departure and LC50. This was also the case for
comparison of the DIBP and DBP LC50 values in N. spinipes. Ecotoxicological similarity for a soil
invertebrate hazard read-across is inferred by the aquatic and benthic invertebrate toxicity comparisons
made between DIBP and DBP, although this inference has slightly greater uncertainty than when it was
made in a previous read-across (U.S. EPA. 2024g). The greater uncertainty is due to a lack to DIBP
sediment exposure data with which to compare to DBP sediment exposure data as a more relevant
ecotoxicological comparison for a soil invertebrate hazard read-across. Because of a lack of predictive
tools to assess ecotoxicological similarity in terrestrial plants and birds, the read-across for these two
taxa was based largely on the physical, chemical, environmental fate and transport agreement between
DIBP and DBP as well as their close structural similarity. Bioconcentration in terrestrial plants was very
similar between DIBP and DBP which increased confidence that both phthalates would behave similarly
in terrestrial plants. Ecotoxicological similarity in algae also helped support the read-across of DBP
terrestrial plant hazard data to DIBP. Uncertainty in establishing ecotoxicological similarity for these
two taxa decreased confidence in the read-across from DBP to DIBP for terrestrial plant and avian
hazard, whereas the aquatic hazard read-across had high confidence, followed by moderate confidence
in the benthic and soil invertebrate hazard read-across from DBP to DIBP. Looking across the multiple
lines of evidence (structural, physical/chemical, ecotoxicological), DBP is an appropriate analog with
high and medium quality aquatic, benthic, and terrestrial hazard data to be used in a read-across to
DIBP.
Page 41 of 54
-------
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
PUBLIC RELEASE DRAFT
December 2024
Appendix B Species Sensitivity Distribution for Acute Aquatic Hazard
The SSD Toolbox (vl.l) is a resource created by EPA's Office of Research and Development
(ORD) that can fit SSDs to environmental hazard data (Etterson, 2020). It runs on Matlab 2018b
(9.5) for Windows 64 bit. For this DIBP risk evaluation, EPA created one SSD with the SSD
Toolbox to evaluate acute aquatic vertebrate and invertebrate toxicity. The use of this
probabilistic approach increases confidence in the hazard threshold identification as it is a more
data-driven way of accounting for uncertainty. For the acute SSD, acute exposure hazard data for
aquatic vertebrates and invertebrates were curated to prioritize study quality and to assure
comparability between toxicity values. For example, the empirical data set included only LC50s
for high and medium quality acute duration assays that measured mortality for aquatic
vertebrates and invertebrates.
Table Apx B-l shows the empirical data that were used in the SSD. To further improve the fit and
representativeness of the SSD, Web-ICE acute toxicity predictions for 72 additional species were added
(Appendix B). Hazard predictions were limited to at or below the limit of water solubility for DIBP (6.2
mg/L). With this data set, the SSD Toolbox was used to apply a variety of algorithms to fit and visualize
SSDs with different distributions.
The Web-ICE application was developed by EPA and collaborators to provide interspecies extrapolation
models for acute toxicity (Raimondo. 2010). These models estimate the acute toxicity (LC50/LD50) of a
chemical to a species, genus, or family with no test data (the predicted taxon) from the known toxicity of
the chemical to a species with test data (the commonly tested surrogate species). Web-ICE models are
log-linear least square regressions of the relationship between surrogate and predicted taxon based on a
database of acute toxicity values. The model returns median effect or lethal water concentrations for
aquatic species (EC50/LC50). Separate acute toxicity databases are maintained for aquatic animals
(vertebrates and invertebrates), aquatic plants (algae), and terrestrial wildlife (birds and mammals), with
1,440 models for aquatic taxa and 852 models for wildlife taxa in Web-ICE version 3.3 (Willming et al..
2016). Open-ended toxicity values (i.e., >100 mg/kg or <100 mg/kg) and duplicate records among
multiple sources are not included in any of the databases. The aquatic animal database within Web-ICE
is composed of 48- or 96-hour EC50/LC50 values based on death or immobility. This database is
described in detail in the Aquatic Database Documentation found on the Download Model Data page of
Web-ICE and describes the data sources, normalization, and quality and standardization criteria (e.g.,
data filters) for data used in the models. Data used in model development adhered to standard acute
toxicity test condition requirements of the ASTM International (ASTM. 2014) and the U.S. EPA Office
of Chemical Safety and Pollution Prevention (e.g. (U.S. EPA. 2016a)).
EPA used empirical DIBP data for the harpacticoid copepod and the fathead minnow and DBP data for
bluegill, opossum shrimp, rainbow trout, zebrafish, the midge (Paratanytarsusparthenogeneticus and
Chironomusplumosus), and the water flea as surrogate species to predict LC50 toxicity values using the
Web-ICE application (U.S. EPA. 2024h). The Web-ICE model estimated toxicity values for 72 species.
For model validation, the model results are then screened by the following quality standards to ensure
confidence in the model predictions. If a predicted species did not meet all the quality criteria below, the
species was eliminated from the data set (Willming et al.. 2016).
. High R2 (>0.6)
o The proportion of the data variance that is explained by the model. The closer the R2 value is
to one, the more robust the model is in describing the relationship between the predicted and
surrogate taxa.
Low Mean Square Error (MSE; <0.95)
Page 42 of 54
-------
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
PUBLIC RELEASE DRAFT
December 2024
o An unbiased estimator of the variance of the regression line.
High slope (>0.6)
o The regression coefficient represents the change in log 10 value of the predicted taxon
toxicity for every change in loglO value of the surrogate species toxicity.
Narrow 95 percent confidence intervals
o One order of magnitude between lower and upper limit
The toxicity data were then used to calculate the distribution of species sensitivity through the SSD
toolbox (Etterson, 2020). The SSD Toolbox's output contained several methods for choosing an
appropriate distribution and fitting method, including goodness-of-fit, standard error, and sample-size
corrected Akaike Information Criterion (BICc, (Burnham and Anderson. 2002)). Most P values for
goodness-of-fit were above 0.05, showing no evidence for lack of fit. The distribution and model with
the lowest BICc value, and therefore the best fit for the data was the Metropolis Hastings: Logistic
(Figure Apx B-l)
TableApx B-l. Species Sensitivity Distribution (SSD) Model Input for Acute Exposure Toxicity
in Aquatic Vertebrates and Invertebrates - Empirica
Data
Genus
Species
Acute Toxicity
Value LCso
(Hg/L)
Reference
Americamysis
bahia
612
(EG&G Bionomics, 1984a)
Danio
rerio
630
(Chen et al., 2014)
Lepomis
macrochirus
788
(Adams et al., 1995; EG&G Bionomics,
1983b; Buccafusco et al., 1981)
Oncorynchus
my kiss
1497
(EnviroSvstem, 1991; EG&G Bionomics,
1983a)
Nitocra
spinipes
3000
(Linden et al., 1979)
Daphnia
magna
3443
(Wei et al., 2018; McCarthy and Whitmore,
1985)
Chironomus
plumosus
4648
(Streufort, 1978)
Pimephales
promelas
5300
(Bencic et al., 2024)
Paratcmytarsus
parthenogeneticus
5800
(EG&G Bionomics, 1984b)
Bolded value indicates DIBP empirical data. Unbolded value indicates DBP empirical data.
TableApx B-2. Species Sensitivity Distribution (SSD) Model Input for Acute Exposure Toxicity
Page 43 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
1375 in Aquatic Vertebrates and Invertebrates - Web-ICE Data
Genus
Species
Acute Toxicity Value LCso
(Hg/L)
Gammarus
pseudolimnaeus
333
Gammarus
pseudolimnaeus
3051
Menidia
peninsulae
318
Catostomus
commersonii
537
Menidia
menidia
495
Caecidotea
brevi cauda
611
Caecidotea
brevi cauda
756
Perca
jlavescens
520
Perca
jlavescens
1413
Allorchestes
compressa
2026
Allorchestes
compressa
289
Allorchestes
compressa
2150
Jordcmella
floridae
924
Sander
vitreus
480
Crassostrea
virginica
2036
Crassostrea
virginica
379
Crassostrea
virginica
243
Oncorhynchus
kisutch
1476
Oncorhynchus
kisutch
445
Oncorhynchus
kisutch
2125
Oncorhynchus
clarkii
1746
Oncorhynchus
clarkii
924
Oncorhynchus
clarkii
1480
Salve linus
namaycush
813
Salve linus
namaycush
637
Salve linus
namaycush
1167
Page 44 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Genus
Species
Acute Toxicity Value LCso
(Hg/L)
Salmo
solar
480
Salmo
solar
1394
Lumbri cuius
variegatus
6099
Salve linus
fontinalis
1321
Salve linus
fontinalis
559
Salve linus
fontinalis
1485
Oreochromis
mossambicus
3763
Oreochromis
mossambicus
1579
Oreochromis
niloticus
967
Micropterus
salmoides
766
Micropterus
salmoides
1089
Oncorhynchus
tshawytscha
1779
Simocephalus
serrulatus
1979
Amblema
plicata
846
Cyprinus
carpio
5624
Cyprinus
carpio
1260
Cyprinus
carpio
3020
Acipenser
brevirostrum
1297
Cyprinodon
variegatus
3672
Cyprinodon
variegatus
1224
Cyprinodon
variegatus
2602
Cyprinodon
variegatus
553
Xyrauchen
texanus
2437
Oncorhynchus
gilae
1365
Lasmigona
subviridis
1996
Salmo
trutta
350
Salmo
trutta
1553
Page 45 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Genus
Species
Acute Toxicity Value LCso
(Hg/L)
Poecilia
reticulata
3310
Poecilia
reticulata
1204
Poecilia
reticulata
3199
Menidia
beryllina
908
Ictalurus
punctatus
5022
Ictalurus
punctatus
1244
Ictalurus
punctatus
2585
Ictalurus
punctatus
1252
Megalonaias
nervosa
1505
Lepomis
cyanellus
1279
Lepomis
cyanellus
2890
Lepomis
microlophus
898
Lithobates
catesbeianus
5832
Lithobates
catesbeianus
1131
Lithobates
catesbeianus
4024
Oncorhynchus
nerka
1930
Utterbackia
imbecillis
2619
Carassius
auratus
5103
Carassius
auratus
5103
Carassius
auratus
1143
Ceriodaphnia
dubia
325
Ceriodaphnia
dubia
2227
Thamnocephalus
platyurus
2814
Margaritifera
falcata
1651
Margaritifera
falcata
289
Daphnia
pulex
2582
Branchinecta
lynchi
2834
Page 46 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Genus
Species
Acute Toxicity Value LCso
(Hg/L)
Lampsilis
siliquoidea
2713
Lampsilis
rajinesqueana
2481
Notropis
mekistocholas
3137
Gammarus
fasciatus
2166
Tigriopus
japonicus
2816
Lymnaea
stagnalis
3440
Acartia
clausi
799
Americamysis
bigelowi
638
Americamysis
bigelowi
873
Bidyanus
bidyanus
2642
Capitella
capitata
1804
Chydorus
sphaericus
1441
Cirrhinus
mrigala
3450
Crcmgon
crangon
3961
Danio
rerio
5006
Danio
rerio
393
Danio
rerio
1881
Etheostoma
lepidum
235
Gibelion
catla
5167
Gibelion
catla
1829
Gila
elegans
4283
Hyalella
azteca
67
Hyalella
azteca
598
Leiostomus
xanthurus
1859
Lepidocephalichthys
guntea
4252
Macrobrachium
nipponense
1651
Morone
saxatilis
1974
Page 47 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Genus
Species
Acute Toxicity Value LCso
(|ag/L)
Ortmcmniana
pectorosa
2518
Palaemofiet.es
pugio
1975
Physella
gyrifia
3256
ThymaUus
arcticus
519
Tisbe
battagliai
60
Xenopus
laevis
4017
1376
1377
3j ModelSelection ~ X
Percentile of interest
Model-averaaed HCd:
Model-averaaed SE of HCd:
CV of HCd
BIC Table
Distribution
b»c r
delta BIC
Wt
HCp
SE HCp
1
weibull
1 3658e+03
0
0 8869
229 2285
39 9621
2
logistic
1 3704e+03
4 5970
0 0891
406.4079
64.3851
3
normal
1 3731e+03
72608
0 0235
377 1761
53 4831
1 4
burr
1 3806e+03
14 8341
5.3296e-04
405.5874
4.2952
I 5 [triangular
1 4054e+03
39 6219
2 2085e-09
135 0860
10 0564
! 6
i
gumbel
1 4123e+03
46.4987
7.0927e-11
296 3624
38 9104
1378
1379 FigureApx B-l. SSD Toolbox Model Fit Parameters
1380
5 |
248.5786]
69.3183
0 27886 I
Page 48 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
DIBP SSD, Metropolis Hastings; Logistic
logistic-MH
» HC05
95% CL HC05
_ . Lumtx/cufas vaneaatus
Pafatanytarsus parthenoge/tel&us ^
Pimepnafes pnametes
CAn^fK^iu^/SmSsus ^
_.Crangoo cranqqf
Cjrrhtnus mngaia z
Daphnta magna?
LwMu'c^r'*-
Nitocra si
Uthobaies cateufy
yranchtnecta
Toxicity Value
FigureApx B-2. Species Sensitivity Distribution (SSD) for Acute DIBP Toxicity to Aquatic Vertebrates and Invertebrates
-------
PUBLIC RELEASE DRAFT
December 2024
Appendix C Environmental Hazard Details
C.l Evidence Integration
Data integration includes analysis, synthesis, and integration of information for the draft risk evaluation.
During data integration, EPA considers quality, consistency, relevancy, coherence, and biological
plausibility to make final conclusions regarding the weight of the scientific evidence. As stated in the
Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (U.S.
EPA. 2021a). data integration involves transparently discussing the significant issues, strengths, and
limitations as well as the uncertainties of the reasonably available information and the major points of
interpretation.
The general analytical approaches for integrating evidence for environmental hazard is discussed in
Section 7.4 of the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a).
The organization and approach to integrating hazard evidence is determined by the reasonably available
evidence regarding routes of exposure, exposure media, duration of exposure, taxa, metabolism and
distribution, effects evaluated, the number of studies pertaining to each effect, as well as the results of
the data quality evaluation.
The environmental hazard integration is organized around effects to aquatic and terrestrial organisms as
well as the respective environmental compartments (e.g., pelagic, benthic, soil). Environmental hazard
assessment may be complex based on the considerations of the quantity, relevance, and quality of the
available evidence.
For DIBP, environmental hazard data from toxicology studies identified during systematic review have
used evidence that characterizes apical endpoints; that is, endpoints that could have population-level
effects such as reproduction, growth, and/or mortality. Additionally, mechanistic data that can be linked
to apical endpoints will add to the weight of the scientific evidence supporting hazard thresholds.
C.l.l Weight of the Scientific Evidence
After calculating the hazard thresholds that were carried forward to characterize risk, a narrative
describing the weight of the scientific evidence and uncertainties was completed to support EPA's
decisions. The weight of the scientific evidence fundamentally means that the evidence is weighed (i.e.,
ranked) and weighted (i.e., a piece or set of evidence or uncertainty may have more importance or
influence in the result than another). Based on the weight of the scientific evidence and uncertainties, a
confidence statement was developed that qualitatively ranks (i.e., robust, moderate, slight, or
indeterminate) the confidence in the hazard threshold. The qualitative confidence levels are described
below.
The evidence considerations and criteria detailed within (U.S. EPA. 2021a) guides the application of
strength-of-evidence judgments for environmental hazard effect within a given evidence stream and
were adapted from Table 7-10 of the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a)
EPA used the strength-of-evidence and uncertainties from (U.S. EPA. 2021a) for the hazard assessment
to qualitatively rank the overall confidence using evidence (Table Apx C-l) for environmental hazard.
Confidence levels of robust (+ + +), moderate (+ +), slight (+), or indeterminant are assigned for each
evidence property that corresponds to the evidence considerations (U.S. EPA. 2021a). The rank of the
-------
PUBLIC RELEASE DRAFT
December 2024
Quality of the Database consideration is based on the systematic review overall quality determination
(high, medium, or low) for studies used to calculate the hazard threshold, and whether there are data
gaps in the toxicity data set. Another consideration in the Quality of the Database is the risk of bias (i.e.,
how representative is the study to ecologically relevant endpoints). Additionally, because of the
importance of the studies used for deriving hazard thresholds, the Quality of the Database consideration
may have greater weight than the other individual considerations. The high, medium, and low systematic
review overall quality determinations ranks correspond to the evidence table ranks of robust (+ + +),
moderate (+ +), or slight (+), respectively. The evidence considerations are weighted based on
professional judgment to obtain the overall confidence for each hazard threshold. In other words, the
weights of each evidence property relative to the other properties are dependent on the specifics of the
weight of the scientific evidence and uncertainties that are described in the narrative and may or may not
be equal. Therefore, the overall score is not necessarily a mean or defaulted to the lowest score. The
confidence levels and uncertainty type examples are described below.
Confidence Levels
Robust (+++) confidence suggests thorough understanding of the scientific evidence and
uncertainties. The supporting weight of the scientific evidence outweighs the uncertainties to the
point where it is unlikely that the uncertainties could have a significant effect on the exposure or
hazard estimate.
Moderate (++) confidence suggests some understanding of the scientific evidence and
uncertainties. The supporting scientific evidence weighed against the uncertainties is reasonably
adequate to characterize exposure or hazard estimates.
Slight (+) confidence is assigned when the weight of the scientific evidence may not be adequate
to characterize the scenario, and when the assessor is making the best scientific assessment
possible in the absence of complete information. There are additional uncertainties that may need
to be considered.
C.1.2 Data Integration Considerations Applied to Aquatic and Terrestrial Hazard
Representing the DIBP Environmental Hazard Database
Types of Uncertainties
The following uncertainties may be relevant to one or more of the weight of scientific evidence
considerations listed above and will be integrated into that property's rank in the evidence table:
Scenario Uncertainty: Uncertainty regarding missing or incomplete information needed to fully
define the exposure and dose.
o The sources of scenario uncertainty include descriptive errors, aggregation errors, errors
in professional judgment, and incomplete analysis.
Parameter Uncertainty: Uncertainty regarding some parameter.
o Sources of parameter uncertainty include measurement errors, sampling errors,
variability, and use of generic or surrogate data.
Model Uncertainty: Uncertainty regarding gaps in scientific theory required to make predictions
on the basis of causal inferences.
o Modeling assumptions may be simplified representations of reality.
Table Apx C-l summarizes the weight of the scientific evidence and uncertainties, while increasing
transparency on how EPA arrived at the overall confidence level for each exposure hazard threshold.
Symbols are used to provide a visual overview of the confidence in the body of evidence, while de-
emphasizing an individual ranking that may give the impression that ranks are cumulative (e.g., ranks of
different categories may have different weights).
Page 51 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
Page 52 of 54
-------
PUBLIC RELEASE DRAFT
December 2024
TableApx C-l. Considerations that Inform Evaluations of the Strength of the Evidence within an Evidence Stream Apical
Endpoints, Mechanistic, or Field Studies)
Consideration
Increased Evidence Strength (of the Apical
Endpoints, Mechanistic, or Field Studies
Evidence)
Decreased Evidence Strength (of the Apical Endpoints, Mechanistic, or
Field Studies Evidence)
The evidence considerations and criteria laid out here guide the application of strength-of-evidence judgments for an outcome or environmental hazard effect
within a given evidence stream. Evidence integration or synthesis results that do not warrant an increase or decrease in evidence strength for a given
consideration are considered "neutral" and are not described in this table (and, in general, are captured in the assessment-specific evidence profile tables).
Quality of the database'1
(risk of bias)
A large evidence base of high- or mediiim-qudXity
studies increases strength.
Strength increases if relevant species are
represented in a database.
An evidence base of mostly /ow-quality studies decreases strength.
Strength also decreases if the database has data gaps for relevant species,
i.e., a trophic level that is not represented.
Decisions to increase strength for other considerations in this table should
generally not be made if there are serious concerns for risk of bias; in other
words, all the other considerations in this table are dependent upon the
quality of the database.
Consistency
Similarity of findings for a given outcome (e.g., of a
similar magnitude, direction) across independent
studies or experiments increases strength,
particularly when consistency is observed across
species, life stage, sex, wildlife populations, and
across or within aquatic and terrestrial exposure
pathways.
Unexplained inconsistency (i.e., conflicting evidence; see U.S. EPA (2005b)
decreases strength.)
Strength should not be decreased if discrepant findings can be reasonably
explained by study confidence conclusions; variation in population or
species, sex, or life stage; frequency of exposure (e.g., intermittent or
continuous); exposure levels (low or high); or exposure duration.
Strength (effect magnitude)
and precision
Evidence of a large magnitude effect (considered
either within or across studies) can increase strength.
Effects of a concerning rarity or severity can also
increase strength, even if they are of a small
magnitude.
Precise results from individual studies or across the
set of studies increases strength, noting that
biological significance is prioritized over statistical
significance.
Use of probabilistic model (e.g., Web-ICE, SSD)
may increase strength.
Strength may be decreased if effect sizes that are small in magnitude are
concluded not to be biologically significant, or if there are only a few
studies with imprecise results.
Biological gradient/dose-
response
Evidence of dose-response increases strength.
Dose-response may be demonstrated across studies
or within studies and it can be dose- or duration-
dependent.
A lack of dose-response when expected based on biological understanding
and having a wide range of doses/exposures evaluated in the evidence base
can decrease strength.
-------
PUBLIC RELEASE DRAFT
December 2024
Consideration
Increased Evidence Strength (of the Apical
Endpoints, Mechanistic, or Field Studies
Evidence)
Decreased Evidence Strength (of the Apical Endpoints, Mechanistic, or
Field Studies Evidence)
Dose response may not be a monotonic dose-
response (monotonicity should not necessarily be
expected, e.g., different outcomes may be expected
at low vs. high doses due to activation of different
mechanistic pathways or induction of systemic
toxicity at very high doses).
Decreases in a response after cessation of exposure
(e.g., return to baseline fecundity) also may increase
strength by increasing certainty in a relationship
between exposure and outcome (this particularly
applicable to field studies).
In experimental studies, strength may be decreased when effects resolve
under certain experimental conditions (e.g., rapid reversibility after removal
of exposure).
However, many reversible effects are of high concern. Deciding between
these situations is informed by factors such as the toxicokinetics of the
chemical and the conditions of exposure, see (U.S. EPA, 1998). cndooint
severity, judgments regarding the potential for delayed or secondary effects,
as well as the exposure context focus of the assessment (e.g., addressing
intermittent or short-term exposures).
In rare cases, and typically only in toxicology studies, the magnitude of
effects at a given exposure level might decrease with longer exposures (e.g.,
due to tolerance or acclimation).
Like the discussion of reversibility above, a decision about whether this
decreases evidence strength depends on the exposure context focus of the
assessment and other factors.
If the data are not adequate to evaluate a dose-response pattern, then
strength is neither increased nor decreased.
Biological relevance
Effects observed in different populations or
representative species suggesting that the effect is
likely relevant to the population or representative
species of interest (e.g., correspondence among the
taxa, life stages, and processes measured or observed
and the assessment endpoint).
An effect observed only in a specific population or species without a clear
analogy to the population or representative species of interest decreases
strength.
Physical/chemical relevance
Correspondence between the substance tested and the
substance constituting the stressor of concern.
The substance tested is an analog of the chemical of interest or a mixture of
chemicals which include other chemicals besides the chemical of interest.
Environmental relevance
Correspondence between test conditions and
conditions in the region of concern.
The test is conducted using conditions that would not occur in the
environment.
" Database refers to the entire data set of studies integrated in the environmental hazard assessment and used to inform the strength of the evidence. In this context,
database does not refer to a computer database that stores aggregations of data records such as the ECOTOX Knowledgebase.
Page 54 of 54
------- |