United States EPA820-R-13-010
Environmental Protection August 2013
Agency
Human Health Benchmarks for Pesticides:
Updated 2013 Technical Document
Introduction
On March 22, 2010, former EPA Administrator Lisa P. Jackson announced a drinking water strategy1 that
outlined four principles to expand public health protection. One of these principles proposed using the
authority of multiple statutes to help protect drinking water. Working together, in 2012, EPA's pesticide
registration and drinking water programs first published Human Health Benchmarks for Pesticides
(HHBPs) for non-carcinogenic effects of 352 pesticides registered for use on food crops. At that time,
EPA committed to updating the HHBPs on a regular basis to reflect the latest scientific information. This
technical document has been developed to support the first such HHBP update. In this update, EPA is
adding 11 new benchmarks for a total of 363 HHBPs, and revising 10 of the HHBPs published in 2012 to
reflect new scientific information. This update also includes benchmarks for 40 pesticides that require
quantification of carcinogenic risk using a linear approach [i.e., cancer slope factor (CSF)].
The HHBPs are advisory values only, and are not legally enforceable federal standards. They are intended
to provide states, tribes, the public and other stakeholders with information to support their internal
decision-making processes (e.g., assist in the interpretation of the drinking water monitoring data).
Derivation of HHBPs
EPA derived the HHBPs by applying the health effects data from pesticide registrations under the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA) and tolerances under the Federal Food, Drug, and
Cosmetic Act (FFDCA) as amended by the Food Quality Protection Act (FQPA), to the typical methods
used for developing drinking water health advisories (HAs) under the Safe Drinking Water Act (SDWA).
Pesticides that have existing HAs or National Primary Drinking Water Regulations (Maximum
Contaminant Levels, MCLs and Maximum Contaminant Level Goals, MCLGs)2 are not included in the
HHBP table.
HHBPs have been developed based on EPA's HA methodology2 combined with RfDs and/or CSFs
developed from health effects data during the pesticide registration process. The HHBPs include only
active ingredients unless metabolites were assessed with the parent compound.. Inert compounds used in
pesticide formulations were not included in this update. HHBPs have been developed for acute (one-day),
chronic (non-cancer), and carcinogenic effects (10~6 - 10"4 risk level) to protect against adverse health
effects from exposure to pesticides that may be found in surface or ground water used for drinking. The
HHBP table lists the acute as well as chronic RfD, the noncancer benchmarks for the sensitive
population/life stage and, when appropriate, the CSF and the corresponding carcinogenic benchmarks. The
acute reference doses (aRfD) are usually determined for children, general population or females of
reproductive age. The chronic reference dose (cRfD) are usually derived for general population or
1 EPA's drinking water strategy can be viewed online at
http://water.epa.gov/lawsregs/rulesregs/sdwa/dwstrategv/index.cfm.
2 These HAs (one-day, ten-day, lifetime) and regulatory standards for drinking water contaminants, including some
pesticides, can be obtained online at: http://water.epa.gov/drink/standards/hascience.cfm.
-------
females of reproductive age. The documentation supporting the RfD and/or CSF derivation for the
specific pesticide is provided by clicking the name of the pesticide from the benchmarks website.
To develop RfDs, EPA examines the entire toxicity database for a pesticide and from this determines the
appropriate studies and endpoints for the acute and chronic dietary risk assessments. EPA's pesticide risk
assessment documents contain a detailed explanation of the basis for establishing the RfDs. If the toxicity
database indicates that toxic effects can be observed following a single oral dose, an aRfD will be
selected. Acute RfDs established for the general population based on systemic/target organ toxicity are
typically also relevant for infants and children and so are also suitable for deriving the one-day HHBP.
Acute RfDs established for females of reproductive age are often based on developmental and
reproductive toxicity which are not appropriate endpoints for deriving HHBPs for children because they
do not represent an effect relevant to that life stage. With regard to the cRfD, in most cases it is
determined for the general population although in rare situations, EPA derives specific cRfD for females
of reproductive age. The same sensitive population is then used for the chronic HHBP. In general, the
methodology to derive the RfDs for pesticide registration is similar to that used to derive HAs for
drinking water (i.e., the same EPA guidance3 is used for reference dose determination).
The HHBPs are derived based on the same default exposure assumptions2 which are used to derive the
HAs for drinking water. When more than one RfD is available for target populations (e.g., aRfD for
general population, females of reproductive age, or children), the aRfD that provides the most health
protective drinking water benchmark will be selected while taking exposure assumptions into
consideration. Since children consume more drinking water per bodyweight basis compared to adults, the
aRfD derived specifically for children would be most appropriate for deriving the acute drinking water
benchmarks. However, since this age-specific value is often not available, the aRfD for general
population is used as a surrogate for children. In such situations, the application of children-specific
exposure assumptions yields the health protective acute benchmarks. For chronic benchmarks, the cRfD
derived for general population or females of reproductive age form the basis for deriving the chronic
drinking water benchmarks.
To derive the acute (one-day) HHBP for children, EPA assumes that the exposed child has a body weight
of 10 kg and consumes 1L of water daily. For deriving the one-day HHBP for females of reproductive
age, a default body weight of 66 kg and daily water consumption of 2L are used. The exposure
assumptions for females represent that for the non pregnant women of reproductive age. For deriving the
lifetime (chronic) HHBP, a body weight of 70 kg (general population) or 66 kg (females 13-49 years),
and a daily water consumption of 2L (for both general population and females 13-49 years) are used.
When there is a cancer concern for a pesticide that requires quantification and the linear approach is
deemed appropriate, a CSF is assigned. Most pesticides do not have a CSF (e.g., non-carcinogens or
threshold type carcinogens). In the case of threshold type carcinogens, the chronic RfDs (and calculated
HHBPs) are considered protective of cancer effects because there is sufficient data for the chemical to
show that carcinogenic effects are not likely to occur below this defined dose (i.e., RfD or HHBP). For
complete details for each chemical, consult the supporting documentation link in the HHBP table.
For pesticide registrations under FIFRA, if there are concerns identified for exposure and/or residual
uncertainties for pre and/or post natal toxicity in the toxicity database, EPA derives acute or chronic
population adjusted doses (PADs) using a FQPA Safety Factor mandated by the FQPA. In the majority of
instances, the PAD and the RfD are the same. It is only in those few instances when the FQPA Safety
Factor is attributed to residual uncertainty with regard to exposure or pre/post natal toxicity that the RfD
3USEPA. 2002. A Review of Reference Dose and Reference Concentration Processes. EPA/630/P-02/002F. This
document can be accessed online at htto://www.epa.gov/raf/publications/pdfs/rfd-fmal.pdf.
-------
and PAD differ. The HA methodology under SDWA does not include this factor. Therefore, HHBP
values were calculated using RfDs, as they are for HAs. Pesticides for which an FQPA Safety Factor was
applied to account for residual uncertainty during the pesticide registration process are identified in the
FiHBP table and footnotes are provided to indicate the magnitude of the factor applied (e.g., 3 or 10).
The Relative Source Contribution (RSC) of 20% is used for derivation of chronic non-cancer FiHBPs and
is a conservative assumption used in EPA's drinking water F£A methodology. The RSC refers to the
percentage of the RfD remaining for drinking water after other sources of exposure to a contaminant are
considered (e.g., diet). Consistent with EPA policies and procedures, the RSC is used only in deriving
chronic FiHBPs; it is not used in deriving acute or carcinogenic HHBPs.
The formulas for determining the acute and chronic HHBPs are presented below:
Acute or one-day HHBP (for children) (ppb) =
aRfD mg/kg BW /day x 10 kg BW x 1000 ((ig/mg)
1L
Acute or one-day HHBP (females 13-49 years) (ppb) =
aRfD mg/kg BW /day x 66 kg BW x 1000 ((ig/mg)
2L
Chronic non-cancer HHBP (general population) (ppb) =
cRfD mg/kg BW /day x 70 kg BW x 1000 (jig/mg) x 0.2 (RSC)
2 L
Chronic non-cancer HHBP (females 13-49 years) (ppb) =
cRfD mg/kg BW /day x 66 kg BW x 1000 (jig/mg) x 0.2 (RSC)
2L
For pesticides that mediate cancer effects via linear dose responses, mathematical models are used to
estimate an upper-bound excess cancer risk associated with lifetime oral exposure. The data used in these
estimates usually come from lifetime exposure studies in animals. EPA generally uses the linearized
multistage model for its quantitative cancer risk assessment. This model fits linear dose-response curves
to low doses and is consistent with a no-threshold model of carcinogenesis, i.e., exposure to even a very
small amount of the substance produces a finite increased risk of cancer.
The linearized multistage model uses dose-response data from the most appropriate carcinogenic study to
calculate a CSF for humans. The CSF is then used to determine the concentrations of the chemical in
drinking water that are associated with theoretical upper-bound excess lifetime cancer risks (1 in
1,000,000 to 1 in 10,000) over a lifetime of exposure.
A two step process is applied to determine the benchmarks specific for cancer effects as in HA
methodology2. In the first step, a drinking water unit risk is determined. In the second step, the drinking
water unit risk is translated to the 10"6 to 10"4 (1 in 1,000,000 to 1 in 10,000) cancer risk levels in water.
The following formulae are applied to estimate the drinking water unit risk and subsequently, to derive
the 10"6 to 10"4 cancer risk levels.
-------
CSF (mg/kg-day)-1 x 2 (L/day) (Adult DW Intake)
Drinking Water Unit Risk ((ig/L"1 or ppb) =
70 (kg) (Adult BW) x 1000 (jig/mg)
From the drinking water unit risk, the following 10"6 to 10"4 cancer risk specific levels in water are
determined.
10-6 or 10-4
10'6 or 10'4 Risk Level in Drinking Water (ppb) =
Drinking Water Unit Risk (ppb)"1
For some carcinogens (e.g., threshold type carcinogens), an RfD or Margin of Exposure (MOE) approach
may be considered protective of cancer risk and, therefore, no separate drinking water cancer risk levels
are needed. If the pesticides are determined to not have carcinogenic potential to humans, or there is
inadequate evidence to determine carcinogenic potential, no separate drinking water cancer risk levels
were determined.
How to View the HHBPs
To view the table of HHBPs and supporting information online go to http://www.epa.gov/pesticides/hhbp.
For More Information
For more information regarding the derivation of HHBPs, contact Santhini Ramasamy in EPA's Office of
Water at ramasamv.santhini@epa.gov.
For information regarding the documentation for deriving the reference doses or cancer risk estimation,
contact Brenda May in EPA's Office of Pesticide Programs at may .brenda@epa.gov.
Abbreviations
aRfD - Acute Reference Dose
BW - Body Weight
cRfD - chronic Reference Dose
EPA - Environmental Protection Agency
FIFRA - Federal Insecticide, Fungicide, and Rodenticide Act
FFDCA - Federal Food, Drug, and Cosmetic Act
FQPA - Food Quality Protection Act
HA- Health Advisory
HHBPs - Human Health Benchmarks for Pesticides
MCL - Maximum Contaminant Level
MCLG - Maximum Contaminant Level Goal
MOE - Margin of Exposure
PAD - Population Adjusted Dose
ppb - parts per billion
CSF- Cancer Slope Factor
RfD - Reference Dose
RSC - Relative Source Contribution
SDWA - Safe Drinking Water Act
------- |