Graduation Year

2005

Document Type

Thesis

Degree

M.S.P.H.

Degree Granting Department

Public Health

Major Professor

Yiliang Zhu, Ph.D.

Committee Member

Getachew Dagne, Ph.D.

Committee Member

Jeff Gift, Ph.D.

Keywords

Dose-response, Biphasic, Neurotoxicity, Risk, Toxicology

Abstract

In collaboration with the United States Environmental Protection Agency (USEPA), the University of South Florida Health Risk Methodology Group has developed dose-time-response models to characterize neurobehavioral response to chemical exposure. The application of dose-time-response models to neurobehavioral screening tests on laboratory animals allows for benchmark dose estimation to establish exposure limits in environmental risk assessment. This thesis has advanced dose-time-response modeling by generalizing a published toxico diffusion model to allow for dose dependent time of peak effects. To accomplish this, a biphasic model was developed which adopted the effect compartment model paradigm used in pharmacokinetics/pharmacodynamics to estimate a distributional rate constant to account for dose related variation in the time of peak effect. The biphasic model was able to describe dose-dependent time of peak effects as observed in the data on acute exposure to parathion and adequately predicted the observed response. However, the experimental design appeared insufficient in statistical power to confirm statistical significance for each parameter of interest. Motivated by the question of what design requirement might be necessary to validate the biphasic model, Monte Carlo simulation was adopted. Simulations were performed to assess the efficacy and efficiency of various experimental designs for detecting and evaluating some critical characteristics of the biphasic model, including the TOPE. The results of simulation suggest that the location of measurement times around the TOPE have important implications for assessing the statistical significance of the parameter that describes dose-dependent TOPE and that the mean squared error of the parameter estimator was improved most when testing times were chosen to bracket the TOPE. While dose dependent time of peak effects has underlying physiological mechanisms such as synergistic or capacity limited kinetics, the biphasic model estimates these physiological properties through a mathematical function which may be physiologically relevant but does not necessarily define physiological mechanisms underlying the response. However, if verified through further testing, the biphasic model may contribute to the USEPA’s aim of developing physiologically relevant dose-response models for assessing risk of neurotoxicity with repeated measurements of response.

Share

COinS