Date of Award

December 2012

Degree Type


Degree Name

Doctor of Philosophy


Biological Sciences

First Advisor

Timothy Ehlinger

Committee Members

Michael Carvan, Christine Cronk, Rienhold Hutz, John Berges


Bioassay, Chemicals of Mutual Concern, Effect Directed Analysis, Risk Assessment, Zebrafish


Multiple interacting stressors in the environment present increasingly complex risks to human health. Too often, however, the data required for traditional risk assessment are either lacking or unavailable at the necessary spatial or temporal scale. In addition, assessment practices and management policies need to move away from single factor approaches in order to accommodate the reality of complex chemical mixtures and environmental stressors. Recent literature suggests that a paradigm shift is under way. This points to a need for the development of new techniques both for rapid data collection and flexible risk assessment strategies that can adapt to make use of readily available data. This dissertation presents two types of methods for improving the risk assessment process given these evolving challenges: predictive analytics and integrated effect-directed toxicity screening.

The first technique addresses the characterization of environmental health using toxicological screening tools. Environmental influences on ecological and human health are often studied using indicators that represent important risk components such as chemical contamination, hazards, exposures, and biological stress. Unfortunately, studies are frequently constrained by the lack of calibrated indicators constructed from standardized metrics.

The second technique is a novel method for population-level risk assessment that uses self-organizing feature maps (SOM) to generate multivariate clusters of cause-of-death and birth outcome metrics, in combination with the use of and supervised learning risk-propagation modelling to evaluate predictability of available indicators. I apply this method to identify exposure-outcome linkages at the county level for Wisconsin, USA and civil divisions in Dobrogea, Romania; thereby providing a dynamic visualization of public health risk relationships with behavioral risk factors (e.g. smoking, heavy drinking) and environmental factors (e.g. land cover, nitrates and faecal coliform in drinking water). These risk relationships do not demonstrate cause-effect, but provide guidance for targeted investigations and for risk-management prioritization.

To investigate a unique way of measuring environmental health, a sediment contact assay using zebrafish (Danio rerio) embryos was adapted from Hollert et al. (2003) as an indicator of teratogenic stress within river sediments. Sediment samples were collected from Lake Michigan tributary watersheds. Sediment contact assay responses were then compared to prevalence of congenital heart disease (CHD) and vital statistic birth indicators aggregated from civil divisions associated with these same watersheds. Significant risk relationships were detected between variation in early life-stage (ELS) endpoints of zebrafish embryos 72 hour post-fertilization and the birth prevalence of human congenital heart disease and infant mortality. Examination of principal components of ELS endpoints suggests that variance related to zebrafish embryonic heart and circulatory malformations is most closely associated with human CHD prevalence.

This study demonstrates a novel application of effect-based toxicity testing for ecological and human health risk assessments. These results support the hypothesis that bioassays normally used for ecological screening can be useful as indicators of environmental stress to humans so as to expand our understanding of environmental - human health linkages. Finally, next steps and new directions for these lines of thinking are discussed.