Risk Characterization

In order to reach the best possible decision in characterizing a particular substance as 'hazardous', all of the available toxicological and epidemiological data should be carefully evaluated. Consideration should be given to the quality of data, the biological relevance of the health parameters monitored, the consistency of the results between studies, and the magnitude of the effects induced. The overall strength of the data implicating a given material as hazardous may then be assessed using a weight of evidence approach.

Assessing the "weight-of-evidence" to characterize the risk posed by a potential toxicant can be addressed in a variety of ways. One approach is based solely on expert judgment in which an individual reflects on the data and offers an informed, yet personal, opinion. A very different approach requires more formal and mathematical procedures such as Bayesian analysis in which data are viewed sequentially and used to formulate a priori and a posteriori judgments. An intermediate approach is one in which a group debates the available data, presents alternative arguments, and collectively reaches a judgment. The EM-COM website has developed a simple framework for evaluating the 'weight of evidence' to characterize a substance as being toxic to the endocrine system.

As discussed in the previous sections, identification and classification of endocrine toxicants has proved challenging. Potential endocrine toxicants comprise many different chemical classes and thus, risk characterization should be determined for each individual toxicant. In general, there is insufficient evidence to fully characterize the risks posed to human health by any toxicant referred to as an 'endocrine disrupter'. This does not negate the importance of rigorous testing and evaluation to determine the properties, mechanisms of action and biological importance of these putative toxicants. Key areas for development include:

• development of appropriate animal models
• critical windows of exposure (timing of exposure)
• measurement of effects at low, environmentally relevant dosages
• identification of mechanisms of action
• global pooling of epidemiological data and the establishment of national and international disease databases
• enhanced cooperation and collaborations between investigators studying effects in human and wildlife populations
• characterization of chemical mixtures and their potential to act as endocrine toxicants
• identification of highly susceptible members of the population to the effects of endocrine toxicants
• characterization of gene-environment factors
• fundamental understanding of normal physiological of the endocrine system in both humans and wildlife species

Other steps in risk assessment consist of: hazard identification, dose-response assessment, and exposure assessment.