Comparison of Four Species Sensitivity Distribution Methods to Calculate Predicted No Effect Concentrations for Bisphenol A

ABSTRACT

Bisphenol A (BPA, CAS RN 80-05-7) is a high production volume chemical used as an intermediate in the production of polycarbonate plastic and epoxy resins. During its manufacture and use, some emissions to surface waters are anticipated. Chronic predicted no effect concentrations (PNECs) for aquatic systems are used to support the assessment of potential risks to aquatic organisms in receiving waters. PNECs for a compound are considered protective of populations, communities, and ecosystems. Traditionally, PNECs are derived by taking the lowest no-observed effect concentration (NOEC) from a set of toxicity studies and dividing by an assessment factor (e.g., 10 to 1000). This traditional approach is appropriate for substances with few data, but may not be necessary for substances with many valid studies. For well-studied substances, statistical approaches (i.e., development of Species Sensitivity Distribution or SSD methods) can be used to calculate a PNEC that makes use of the full distribution of available NOEC values. Bisphenol A has an extensive set of aquatic toxicity studies covering diverse taxa including algae, hydra, rotifers, mollusks, crustaceans (both benthic and pelagic), insects, annelids, fish, and amphibians. The full chronic data set was used to calculate PNEC values using four SSD methods: (1) the Hazard Concentration (HC5) approach developed by The Netherlands National Institute of Public Health and the Environment (RIVM), (2) the U.S. Environmental Protection Agency's water quality criteria procedure, (3) SigmaPlot (Systat 2000) commercial software that calculates percentile values, and (4) a distributional method consistent with that used by Environment Canada. Using these approaches, PNEC values for BPA range from 11 to 71 μ g/L. Literature studies suggest that application of an additional assessment factor is unwarranted if an SSD-based PNEC is based on chronic data. SSD-derived PNEC values and the traditionally derived PNEC value of 1.6 μ g/L are then compared to concentrations of BPA that have been measured in North American and European surface waters. Adverse risks to aquatic organisms are not anticipated from measured concentrations of BPA in North American and European surface waters.

Key Words:

• no observed effect concentration (NOEC)
• Final Chronic Value (FCV)
• hazard concentration (HC5)
• hazard quotient (HQ)
• PNEC

ACKNOWLEDGMENT

This work was sponsored by the Polycarbonate/BPA Global Group, Arlington, Virginia.

Notes

*All studies used to derive the PNECs have been screened for study quality according to the procedures described in Staples et al. (2002). Concentrations presented here are either nominal or mean measured as reported in each study.

**The longest duration medaka study was used for this species (NOEC = 247 μ g/L).

*Hazard quotients calculated as the ratio of the surface water concentration divided by the range of SSD-based PNEC values (11 to 71 μ g/L).

**Represents only the number of unique sampling locations, eliminating duplicate samples and multiple sampling events at any given location.

***Varying detection limits.