How should risk assessors integrate different types of scientific information (animal studies, human studies, literature reviews, computer models etc.), diverging scientific information or information emerging from new methods? How confident are scientists about the results of their conclusions given the inherent uncertainties related to any scientific assessment? The growing size and number of datasets and new forms of evidence are driving the need for clear and consistent approaches to weighing evidence and assessing uncertainties, which produce more robust assessments in support of decision-making.
Recent developments, future directions
Demands to weigh evidence more systematically and transparently in part stem from the needs for more complex assessments (e.g. endocrine activity, chemical mixtures) while integrating new sources of evidence (e.g. in vitro, omics etc.), explained Lorenz Rhomberg, a leading global expert in risk assessment. The global harmonisation, where possible, of the array of methodologies available is a key challenge for the risk assessment and regulatory science communities. This will ensure robustness and consistency of the approaches while remaining fit for purpose and facilitating clear communication to the outside world.
Scientific evidence changes with scientific advances
Over the last 40 years, the revolution in molecular biology has made possible the study of toxicity at the cellular and molecular level to predict possible toxic effects of chemicals in complex organisms. Harvey Clewell of the Center for Human Health Assessment at the Hamner Institutes for Health Sciences in the U.S.A., explored how mode of action analysis is already complementing and in some cases replacing empirical testing in animals, the mainstay of chemical risk assessment for the past century.
Evidence and uncertainty in microbiological risk assessment
Matthias Greiner of the Federal Institute for Risk Assessment (BfR) in Germany and University of Veterinary Medicine Hannover, revealed that in microbiological risk assessment current thinking, terminology and practice on evidence integration and uncertainty assessment differ from chemical risk assessment. He argues convincingly that there is room to make these approaches more systematic to support decision-making on increasingly complex problems.
Evidence and uncertainty in environmental risk assessment
Ecologists have developed many of the key concepts currently used in human risk assessment, stated Glenn Suter of the U.S. Environmental Protection Agency’s National Center for Environmental Assessment. He presents the preliminary results for future EPA guidance for ecological risk assessment dealing with a structured weight of evidence approach that aims to balance quantitative and qualitative evidence with confidence in the results.
Predictive toxicology – new methods create new uncertainties
More complex assessments and demands to reduce animal testing are driving alternative methods for predictive toxicology. Maurice Whelan’s work as head of both the EU Reference Laboratory for Alternatives to Animal Testing (EURL ECVAM) and the Systems Toxicology Unit of the Institute for Health and Consumer Protection (IHCP), of the European Commission's Joint Research Centre (JRC), are at the heart of these developments in Europe. These new developments employing alternatives to traditional toxicological testing methods have major advantages including reducing animal testing and test modes of action to a great extent but bring new types of emerging uncertainties that should be considered in assessments.
Using uncertainty to strengthen scientific assessments
Andrew Hart of the UK Food and Environment Research Agency (FERA) wrapped up the session with an engaging overview of the EFSA Scientific Committee’s draft guidance on uncertainty. The toolbox of methods it proposes is intended to ensure a flexible and consistent approach to uncertainty analysis in all EFSA assessments, including options to include a quantitative assessment of the overall uncertainty.