A framework for integrating evidence to assess hazards and risk

Area of Science:

• Toxicology
• Epidemiology
• Risk Assessment

Background:

• Accurate human health hazard characterization requires integrating human, animal, and mechanistic data.
• Mechanistic data are crucial for linking animal and human studies and assessing relevance and uncertainty.
• Existing frameworks may not fully integrate diverse evidence types for robust hazard assessment.

Purpose of the Study:

• To present a novel evidence integration framework (EIF) for synthesizing multiple lines of scientific evidence.
• To provide a systematic method for assessing human health hazards by integrating epidemiological, toxicological, and mechanistic data.
• To enhance the characterization of chemical relevance and uncertainty in risk assessment.

Main Methods:

• Developed a novel evidence integration framework (EIF) for data synthesis.
• Organized toxicological and epidemiological data based on observed human health effects (disease-based component).
• Organized data based on proposed mechanisms of action (mechanism-based component).
• Included methods for cross-classification, concordance assessment, and uncertainty characterization.

Main Results:

• The EIF organizes data by both health outcomes and mechanisms of action.
• It allows for cross-classification and assessment of data concordance and uncertainty.
• The framework facilitates the identification of knowledge gaps and the impact of uncertainty on causal inference.

Conclusions:

• The novel EIF provides a robust method for synthesizing diverse scientific data for hazard characterization.
• This framework supports evidence-based decision-making in toxicology and risk assessment.
• The EIF improves the understanding of chemical toxicity and human health risks by integrating multiple evidence streams.