You are here:
Assessing Synergy in Toxicology and EpidemiologyEPA Grant Number: FP916367
Title: Assessing Synergy in Toxicology and Epidemiology
Investigators: Howard, Gregory J.
Institution: Boston University
EPA Project Officer: Zambrana, Jose
Project Period: January 1, 2004 through December 31, 2006
Project Amount: $91,933
RFA: STAR Graduate Fellowships (2004) RFA Text | Recipients Lists
Research Category: Fellowship - Public Health Sciences , Academic Fellowships , Health Effects
The dramatic growth in the number of chemicals in widespread use recently has led to increasing concern about the effects of interactions between low-level exposures. In particular, mixtures can sometimes produce dramatically stronger effects (“synergy”) than would be predicted by a naive model. Many general approaches and statistical tests have been proposed to address combination exposures, but there is not universal agreement on their appropriateness. A number of otherwise promising statistical tests are parametric in nature and thus make strong assumptions, not always correct, about the shapes of the individual dose-response curves. The objective of this research is to contribute to an understanding of interactive effects, and particularly of the appropriateness of concentration addition, when applied to difficult situations like the combination of a full with a partial agonist. This will have important benefits for risk assessments of combination exposures.
This project has several goals. First, I will compare the results of algebraic and numerical models of receptor-based systems with the predictions made by concentration addition and other means of estimating interactive effects. In this way, the theoretical underpinnings of the isobole method and of concentration addition, particularly in complex model systems that include multiple agents and partial agonists, will be investigated. Second, I will develop a new statistical test for interaction. Rather than relying on parametric assumptions, techniques will be borrowed from spatial mapping (in particular, generalized additive models) to analyze the entire response surface for interactive effects with a nonparametric test that makes no assumptions about dose-response curves. Third, I am developing experimental methodology for testing synergy through the use of standard laboratory techniques, for example, 96-well plates. The usefulness of various statistical methods with experimental data generated by our laboratory collaborators will be examined. Finally, I hope to extend some of these concepts to epidemiology, in which the shape of the dose-response curve is not generally considered when evaluating interaction. In particular, I hope to develop an analogy for concentration addition for epidemiologic data.