Best Practices for Evaluating Method Validity, Data Quality and Study Reliability of Toxicity Studies for Chemical Hazard and Risk Assessments

Risk assessments on chemicals are intended to provide sufficient information to evaluate the potential hazards associated with exposure to that chemical. It is important that risk assessments be based on the most reliable information available so that they can be used for sound decision-making by regulators and to convey appropriate risk information to the public. Although there are many uncertainties in risk assessments, these can be reduced by ensuring that the risk assessment reflects the most reliable and scientifically supported information on the potential hazards of the chemical, and that the quality and the completeness of the data are at a high standard. There are many different types of studies that a risk assessor needs to evaluate, and these vary not only in study design and methodologies, but also in quality. For the evaluation of the potential hazards to be credible, as well as objective and scientifically supported, a standardized procedure for data evaluation is desirable; this will also ensure consistency and transparency which are key factors in public acceptance. The risk assessor needs a process to ensure that data are of sufficient quality to underlie the hazard and risk assessments. This paper examines existing approaches or procedures used to evaluate data quality from in vivo and in vitro mammalian toxicity studies for use in hazard and risk assessments. Based on this review, the Klimisch approach with an expanded list of narrative descriptions should be used as a best practice for studies that have internationally accepted testing guidelines. The ToxRTool is proposed as the best practice to evaluate data quality for in vitro studies, as well as for studies that do not have internationally accepted guidelines.