The USEPA has changed some of their procedures for calculating various human-health benchmarks since the previous revisions to the HBSL methodology in 2007 (Toccalino, 2007) and 2014 (Toccalino and others, 2014). The HBSL methodology was updated to reflect those changes.

Changes in USEPA Methods for Calculating Benchmarks

In 2017, USEPA began using updated default exposure assumptions based on the National Health and Nutrition Examination Survey (NHANES) data from 1999 to 2006 to calculate Chronic Noncancer and Carcinogenic HHBPs (USEPA, 2017). The default body weight and drinking water intake rate were changed to 80 kg and 2.5 L/day, respectively, from the previous default body weight and intake values of 70 kg and 2 L/day, respectively (USEPA, 2017). Similarly, the default body weight and drinking water intake values used for USEPA’s updated Human Health Ambient Water Quality Criteria were changed to 80 kg and 2.4 L/day, respectively (USEPA, 2015). The 2.5 L/day rate represents the "consumers only" direct and indirect community water ingestion at the 90th percentile for adults ages 21 and older, while the 2.4 L/day rate represents the total population, averaging consumers and non-consumers of tap water (USEPA, 2015 and 2017).

In addition, the 2017 Chronic Noncancer HHBPs use the chronic population adjusted dose (cPAD) instead of the chronic reference dose (cRfD), when different. For most pesticides the cPAD is equivalent to the cRfD; however, for some pesticides an extra safety factor mandated by the Food Quality Protection Act for uncertainty related to pre/post-natal exposure or toxicity or for uncertainty in the overall completeness of the toxicity database is applied to the cRfD to derive a cPAD (USEPA, 2017).

The 2018 HBSL update incorporates USEPA's updated exposure assumptions (80 kg body weight and 2.5 L/day drinking water intake) and, for pesticides, the use of the cPAD when it differs from the cRfD.

Calculation of HBSLs

HBSLs are calculated only for compounds that do not have USEPA Maximum Contaminant Levels (MCLs) or Chronic Noncancer or Carcinogenic HHBPs. Noncancer HBSLs are calculated using the default exposure assumptions that are used to derive USEPA Chronic Noncancer HHBPs. Equation 1 incorporates a drinking-water Relative Source Contribution (RSC) factor that defaults to 20% in the absence of contaminant-specific data. The RSC assumes that 20% of the total exposure from all sources to a contaminant comes from drinking water.

Noncancer HBSL (µg/L)=	(cRfD or cPAD [mg/kg-day]) × (80 kg body weight) × (1,000 µg/mg) × RSC	(1)
	2.5L water consumed/day

The Cancer HBSL range (concentrations representing a 10^-6 to 10-⁴ cancer risk range) is calculated using the same default exposure assumptions that are used to derive USEPA Carcinogenic HHBPs (eq. 2):

Cancer HBSL (µg/L)=	(80 kg body weight) × (risk level)	(2)
	(2.5L water consumed/day) × (SF or Q1* [mg/kg-day]-1) × (mg/1,000 µg)

Risk levels of 10^-6 and 10-⁴ are used to calculate the low end and high end, respectively, of the Cancer HBSL range (eq. 2) using the oral cancer slope factor (SF) or oral carcinogenic potency factor (Q1*). The default exposure assumptions used in equations 1 and 2 represent an adult's lifetime exposure to a contaminant in drinking water. The USGS consults with USEPA when HBSL calculations are not straightforward. There are a few contaminants for which HBSLs were calculated using exceptions to the HBSL methodology that is described above. Learn more.

Sources of Toxicity Information

The human-health toxicity information (cRfD or cPAD values, and SF or Q1* values) used to calculate HBSLs meets all of the following criteria: (1) published by USEPA, (2) peer reviewed, (3) publicly available, and (4) the most recently available. For a given contaminant, toxicity information from the most recent of these four primary USEPA data sources is used:

• Integrated Risk Information System (IRIS) — Office of Research and Development
• Drinking Water Standards and Health Advisories document — Office of Water (OW)
• Human-health risk assessments and other documents from USEPA's pesticide registration and review process — Office of Pesticide Programs (OPP)
• Chemicals Evaluated for Carcinogenic Potential document — OPP

In the absence of available toxicity information from the primary USEPA data sources above, the USEPA Provisional Peer Reviewed Toxicity Values for Superfund (PPRTV) were used, when available. This is consistent with the use of PPRTVs by the Superfund Program as a "second tier" source of toxicity values. PPTRVs are externally peer reviewed and were developed following USEPA guidance on deriving human-health toxicity values for the Superfund Program; however, the toxicity values have not undergone the multi-program consensus review required for inclusion in the USEPA IRIS database (USEPA, 2018b).

This guidance describes how HBSLs are used in USGS water-quality assessments. Measured concentrations of contaminants in surface-water or groundwater sources of drinking water are compared to human-health benchmarks to provide perspective on the potential relevance of detected contaminant concentrations to human health. Concentrations of contaminants that are regulated by USEPA in drinking water under the Safe Drinking Water Act are compared to USEPA MCLs. Concentrations of unregulated contaminants are compared to either USEPA HHBPs or to USGS HBSLs, when available. Learn more.

Comparing Contaminant Concentrations to Benchmarks

USEPA MCLs, USEPA chronic HHBPs, and USGS HBSLs generally are calculated assuming an adult’s lifetime of exposure to drinking water (see eqs. 1 and 2). As a result, contaminant concentrations (or summary statistics for concentrations) indicative of long-term exposure are most appropriate to compare to these benchmarks in most USGS applications (Toccalino, 2007). In groundwater, concentrations measured in individual samples (wells) are compared to these benchmarks. In surface waters, time-weighted annual mean concentrations are used.

Some contaminants have as many as three HBSLs or chronic HHBPs:

• Chronic noncancer
• Low end of cancer risk range (10^-6)
• High end of cancer risk range (^10-4)

For contaminants with multiple benchmarks, data analysts can compare measured concentrations (or summary statistics for concentrations) to benchmarks in water-quality assessments as follows:

1. First compare concentrations or summary statistics to the lowest HHBP or HBSL because this is the most conservative (protective) benchmark. For nearly all carcinogens or possible carcinogens, the lowest of the three values will be the HHBP or HBSL that corresponds to a 10^-6 cancer risk.
2. If concentrations are less than the lowest HHBP or HBSL, then no further benchmark comparisons are needed.
3. If concentrations are greater than the lowest HHBP or HBSL, and the lowest benchmark value is the low end of the cancer risk range (10^-6), then compare concentrations to the high end of the cancer risk range (10^-4) to determine where concentrations fall within the 10^-6 to 10^-4 cancer risk level range. USEPA advises consideration of cancer risk benchmarks that represent a 10^-6 to 10^-4 cancer risk level range, when appropriate (USEPA, 2018a).

Learn more about comparing contaminant concentrations to benchmarks and interpreting the potential human-health significance of detected contaminants.

See Toccalino (2007) for additional information about:

• Exceptions to the HBSL methodology
• Guidance on the use of benchmarks
• Comparisons of contaminant concentrations to benchmarks
• Interpretation of the potential human-health significance of detections
• Limitations of screening-level assessments

See Toccalino (2014) for the 2014 updates to the HBSL methodology, including:

• Adopting HHBPs and removing HBSL values for pesticides with HHBPs
• Calculating noncancer HBSLs for contaminants regardless of carcinogenicity status
• Discontinuing the use of a 10-fold risk management factor for Group C carcinogens

• Lisa Nowell of the USGS performed a quality-control check of updated HBSL values.
• Two USGS colleague reviewers provided insightful and helpful reviews.
• Alice McCarthy from the USGS Office of Water Information, Integrated Information Dissemination Division managed website development for the 2018 HBSL update.

Toccalino, P.L., Nowell, L.H., Wilber, W.G., Zogorski, J.S., Donohue, J.M., Eiden, C.A., Krietzman, S.J., and Post, G.B., 2003, Development of health-based screening levels for use in state- or local-scale water-quality assessments: U.S. Geological Survey Water-Resources Investigations Report 03-4054, 22 p.

Toccalino, P.L., 2007, Development and application of health-based screening levels for use in water-quality assessments: U.S. Geological Survey Scientific Investigations Report 2007-5106, 12 p.

Toccalino, P.L., Norman, J.E., and Schoephoester, K.M., 2014, Health-Based Screening Levels: Updated 2014 technical information. U.S. Geological Survey web page, accessible at https://water.usgs.gov/water-resources/hbsl/methods-guidance-2014.html

USEPA, 2015, Human health ambient water quality criteria: 2015 Update: U.S. Environmental Protection Agency, Office of Water, EPA 820-F-15-001, June 2015, 5 p.

USEPA, 2017, Human health benchmarks for pesticides: Updated 2017 technical document: U.S. Environmental Protection Agency, Office of Water and Office of Pesticide Programs, EPA 822-R-17-001, January 2017, 5 p.

USEPA, 2018a, 2018 Edition of the drinking water standards and health advisories: U.S. Environmental Protection Agency, Office of Water, EPA 822-F-18-001, March 2018, 20 p.

USEPA, 2018b, Provisional Peer Reviewed Toxicity Values for Superfund (PPRTV): U.S. Environmental Protection Agency, Office of Superfund Remediation and Technology Innovation, Accessed April 10, 2018 at https://hhpprtv.ornl.gov/.