Development of Statistical Approaches to Provide Toxicity Estimates from Exposure Data to Support Risk Assessments

Period of Performance: 03/22/2002 - 09/21/2002


Phase 1 SBIR

Recipient Firm

Menzie-cura & Assoc., Inc.
1 Courthouse Lane, Ste 2
Chelmsford, MA 01824
Principal Investigator


The technical objective in Phase I is to review and identify potential statistical approaches for obtaining probabilistic concentration and exposure threshold values for screening contaminants of military interest (military unique compounds, or MUCs) based on existing laboratory-based toxicity and bioaccumulation data. The goal is to develop validated statistical approaches to determine the uncertainty, or confidence limits, surrounding exposure-based (e.g., bioaccumulation) toxicity estimates for use in ecological risk assessments. These tools and approaches will enhance and further develop current ecological risk assessment techniques, which tend to focus on single threshold values derived deterministically, and typically do not provide confidence intervals even in the case of probabilistic assessments. Following the identification, review, and application of potential statistical approaches that may be used to estimate concentration-based toxicity thresholds from existing toxicity and bioaccumulation data sets, we will evaluate the selected approaches. This will be done using chemical data sets that can support development of body-residue toxicity threshold and confidence intervals. We will compare the results of applications of the statistical approaches for the selected contaminants with the goal of developing an approach that combines the best features of existing methods. We anticipate developing an integrated framework to evaluate bioaccumulation and potential toxicity using sophisticated statistical techniques that quantitatively characterize uncertainty. Particular emphasis will be placed on applications of these models in cases with limited toxicity and bioaccumulation data. The statistical models developed here will be valuable for application at U.S. Army installations and training ranges as well as for civilian uses such as environmental evaluations required under the U.S. Army Corps of Engineers (USACE) dredging program. This work will support environmental decisionmaking at every level and inform the regulatory process by providing quantitative uncertainty information for available toxicity and bioaccumulation data. Models and statistical approaches to developing screening levels and guidelines are in great demand by risk assessors, regulators, and the regulated community. Many users of these values are unaware of the inherent uncertainty in the calculations. This product will provide a method for obtaining threshold concentrations and associated confidence levels to characterize uncertainty in support of ecological risk assessments.