A systematic review of methods of uncertainty analysis and their applications in the assessment of chemical exposures, effects, and risks

Abstract

Methods of uncertainty analysis are being included increasingly in regulatory chemical risk assessment. Although best practices have been established by several safety agencies in Europe and the United States, they exist only in the grey literature - there has been no comprehensive analysis of the scientific, peer-reviewed literature on these methods. We therefore conducted a systematic review of the recent peer-reviewed literature (2007-2013) on uncertainty analysis relevant to chemical risks. The main objective was to determine whether current methods are robust enough for regulatory use, because the methods used to protect public health must meet the most stringent scientific standards. Based on 297 papers, we concluded that the peer-reviewed literature is much more critical about the disadvantages of those methods, compared to the grey literature. Furthermore, uncertainty analyses can be influenced significantly by subjective expert judgment. As a suggested improvement, we developed guidelines for transparent reporting of uncertainty assessment results.

Radiation risks and uncertainties: a scoping review to support communication and informed decision-making

“…Our results point out that there is no scientific consensus on definitions of uncertainty. This is not entirely new as it was partly expected based on the arguments of multiple authors that uncertainty definitions are numerous and diverse (Walker et al 2003, Maxim 2014. However, we argue that if it is not possible to use a common/universal definition of uncertainty (given the big differences across disciplines), at least the authors/users of uncertain information have to clearly describe what they mean by the concept of uncertainty.…”

“…As is the case with the other types of uncertainties, decision uncertainty has also been termed in different ways and sometimes even blended with ambiguity, such as in the case of low-dose (ioni sing and non-ioni sing) radiation (Renn 2008b). (Maxim 2014) also argues that there are more types of uncertainties in addition to the 'epistemic-aleatory-ambiguity' distinction. These include technical uncertainties, which are mainly technical errors caused by imprecise instruments or measurement methods; methodological uncertainties which include methodological challenges, such as making assumptions when knowledge is missing or choosing from amongst several available methods for assessing a parameter; normative uncertainties, which include interpretation of raw data and conclusions about the level of evidence they provide; as well as communication uncertainties, which include how completely and understandably the research is reported (Maxim 2014: 5).…”

Comparison of methods for calculating the health costs of endocrine disrupters: a case study on triclosan

“…Further, potential methodological drawbacks should be clarified before using the IPRA method for regulatory ends. Indeed, BMD modelling uses numerous assumptions [ 56 ] which can strongly influence the use of BMD-based methods, including the IPRA method used here, for calculating the share of the population susceptible that shows a negative effect related to exposure to TCS or another chemical. Such analysis is critical given the current increasing tendency to use such models for regulating chemical risks in Europe, where they can be used as “black boxes” that give needed figures but are not understood in their inner structures and assumptions [ 56 ].…”

“…Indeed, probabilistic modeling including BMD can be significantly influenced by subjective expert judgment and assumptions behind the tool itself and its recommended use [ 56 ] e.g., the choice of the 5% as typical level of significance used to choose the best-fitted curve, the criteria used for choosing the best-fitted model (acceptability, similarity with the log-likelihood with the full model), the choice of the BMR (Benchmark Response) of 5% (whereas levels of 1% to 10% can be chosen and have been reported in the literature). Furthermore, the BMD results depend of the sample size of the original studies, i.e.…”