PARENT SESSION
PM16 - Probabilistic Ecological Risk Assessment
Monday, 18 November 2002
8:00 AM to 6:30 PM
Exhibit Hall

(P379) A Comparison of Methods to Conduct a Hazard Assessment of Bisphenol A.

Staples, Charles^*,1, Woodburn, Kent², Caspers, Norbert³, Hall, A. Tilghman⁴, Klecka, Gary², ¹ Assessment Technologies, Inc., Fairfax, VA, USA² Dow Chemical Company, Midland, MI³ Bayer AG, Leverkusen, Germany, Germany⁴ Bayer Corporation, Stilwell, KS

ABSTRACT- Bisphenol A (BPA; 4,4-isopropylidene diphenol) is a chemical intermediate used primarily in the production of epoxy resins and polycarbonate products. BPA has been identified in surface waters and hence, has been the subject of considerable research into its potential effects on aquatic organisms. Available literature on aquatic toxicity of BPA was reviewed for quality against USEPA and OECD GLP principles. From this review, studies of suitable quality covering numerous ecologically relevant endpoints were identified to evaluate the survival, growth and reproductive success of aquatic organisms exposed to BPA. Those studies yielded several dozen no observed effect concentrations (16 to 3640 g/L) and lowest observed effect concentrations (160 to 11,000 g/L) that were considered in this hazard assessment. Across all data, adverse effects on survival, growth and reproduction occurred only at concentrations of 160 g/L and above. Four different approaches are employed to assess the aquatic hazard of BPA: 1) USEPA's Concern Concentrations; 2) Weight of Evidence; 3) USEPA's ambient water quality criteria; and 4) Environment Canada's Tier 3 distributional analysis. The first approach is the most conservative approach and is useful mainly for small databases. It relies on assessment factors. The Weight of Evidence approach relies on an analysis of multiple studies addressing similar endpoints (e.g., reproduction or growth). The third and fourth approaches utilize the full aquatic database to develop a threshold concentration considered protective of aquatic communities and ecosystems by setting the protection level at the 95th percentile. Finally, the four types of hazard values (i.e., toxicity thresholds) are then compared with typical ambient surface water concentrations found in North America. The results show that typical concentrations of BPA in surface waters are below the various toxicity thresholds, but to different degrees. The pluses and minuses of each approach are discussed.

Key words: concern concentrations, weight of evidence, water quality criteria, distributional analysis