Scientific uncertainty

One can identify two major categories of uncertainty in EIA data (scientific) uncertainty inherited in input data (e.g., incomplete or irrelevant baseline information, project characteristics, the misidentification of sources of impacts, as well as secondary, and cumulative impacts) and in impact prediction based on these data (lack of scientific evidence on the nature of affected objects and impacts, the misidentification of source-pathway-receptor relationships, model errors, misuse of proxy data from the analogous contexts) and decision (societal) uncertainty resulting from, e.g., inadequate scoping of impacts, imperfection of impact evaluation (e.g., insufficient provisions for public participation), human factor in formal decision-making (e.g., subjectivity, bias, any kind of pressure on a decision-maker), lack of strategic plans and policies and possible implications of nearby developments (Demidova, 2002). 

Here we come to the heart of the problem we shall explore in this book just how certain is our science on matters such as this And how should public health officials deal with the uncertainties We shall be exploring the two responses to the FDA s position that were set out earlier and learn what we can about their relative scientific merits not specifically in connection with the aflatoxin problem, but in a more general sense. We shall also be illustrating how regulators react to these various scientific responses, and others as well, using some examples where the economic stakes are very high. One would like to believe that the size of the economic stakes would not influence scientific thinking, but it surely influences scientists and policy-makers when they deal with scientific uncertainties. 

As what we have termed the regulatory approach is discussed, it should be kept in mind that much of what is done in the conduct of risk assessment is to ensure a high degree of consistency and predictability in the face of scientific uncertainty. There are many scientists who will disagree with the regulatory approach. Sometimes the disagreement arises in specific cases, where evidence has arisen that seems to question one or more of the typical regulatory defaults. Other scientists may find the whole process, because it is dependent upon untested... 

The Lehman-Fitzhugh approach has been very widely used for setting limits on exposures to chemicals, not only in food, but in all other environmental media, but it has undergone significant refinement in recent years. The EPA, and others, for example, now uses the term uncertainty factor for those factors that reflect a true scientific uncertainty, and distinguishes these from safety factors, which reflect the injection of policy judgments that go beyond scientific uncertainties in the establishment of acceptable intakes. The EPA has dropped the... 

The policy decision to act before science is certain does not, of course, dissolve the scientific uncertainties. Indeed, a strong argument can be made that an assessment of the type we have described should not pretend to represent normal science. Many of its outcomes are untestable with current methods this alone might disqualify it as a true science. 

Application of a UF to the PNAEL to take into account the degree of scientific uncertainty involved. The following degrees of confidence in the human PNAEL are suggested high = 1, medium = 1-2, low = larger UF. 

US-EPA (1993) stated that in addition to the standard factors (for inter- and intraspecies differences, less than chronic duration studies, and LOAEL-to-NOAEL extrapolation), an extra factor should be included if the total toxicological database is incomplete, i.e., the so-called modifying factor (ME). It was stated that the magnitude of the MF depends upon a professional assessment of scientific uncertainties of the study and database not explicitly accounted for by the standard factors, e.g., the completeness of the overall database and the number of species tested. The default value for the MF is 1. 

Baird et al. (1996) suggested a probabilistic alternative to the practice used by the US-EPA to derive RfDs from a NOAEL and application of UFs. The probabilistic approach expresses the human population threshold for a given substance as a probability distribution of values, rather than a single RfD value, taking into account the major sources of scientific uncertainty in such estimates. The approach was illustrated by using much of the same data that US-EPA used to justify their RfD procedure. For the four key extrapolations that were considered necessary to define the human population threshold based on animal data (interspecies, interindividual, LOAEL-to-NOAEL, and subchronic-to-chronic), the proposed approach used available data to define a probability distribution of each adjustment factor, rather than using available data to define point estimates of UFs. 

Communicating in a manner that accurately portrays risk and the nature of confidence in results is integral to and a major challenge for practitioners of probabilistic risk assessment. Inadequate communication of scientific uncertainty about the effect, severity, or prevalence of a hazard tends to increase unease among decision makers, stakeholders, and other participants. Efforts of the risk assessors should provide clarity for decision makers who must in turn bear ultimate responsibility for communicating the parameters of any decision. The risk assessor will work within a framework that must be clearly communicated by decision makers during the problem formulation. Decision makers at the outset of the risk assessment process must articulate the following points ... 

Brennan, Troyen A. 1988. Causal Chains and Statistical Links The Role of Scientific Uncertainty in Hazardous-Substance Litigation. Cornell Law Review 73 (March) ... 

Reproductive risk descriptors are intended to address variability of risk within the population and the overall adverse impact on the population. In particular, differences between high-end and central tendency estimates reflect variability in the population but not the scientific uncertainty inherent in the risk estimates. There is uncertainty in all estimates of risk, including reproductive risk. These uncertainties can result from measurement uncertainties, modelling uncertainties and assumptions made due to incomplete data. Risk assessments should address the impact of each of these uncertainties on confidence in the estimated reproductive risk values. 

One possible variation of Alternative 3 would be to set RMCLs as a range of finite risk levels. This alternative would recognize the lack of accuracy and precision of risk calculations and the inherent difficulties in selecting one finite level as the only appropriate health goal in view of the numerous scientific uncertainties of risk estimates. 

As in other risk-assessment approaches (e.g., NRC 1994), scientific uncertainties are a predictable feature of any new biomonitoring-led risk assessments. As shown above and discussed more fully in Chapter 6, identifying and communicating those uncertainties—such as the effect of interindividual variation in elimination rate and limits on extrapolating adult PK data to children—are critical in communicating the risk results. 

Miles, S., and L.J. Frewer. 2003. Public perception of scientific uncertainty in relation to food hazards. J. Risk Res. 6(3) 267-284. 

Not very long after the discovery of the deficiency of isoniazid acetylation, it turned out that the frequency of this phenotypic fault also differed much between the world s populations (47). Today, the reason for this particular interethnic variability is still unknown an influence of geographical latitude and climate is suspected. Such scientific uncertainty is common in this field of research, and it contrasts with the clear-cut findings relating malaria resistance and G-6-PD deficiency (see below). 

Where there are threats of serious or irreversible damage, scientific uncertainty shall not be used to postpone cost-effective measures to prevent environmental degradation. 

There might be sufficient scientific uncertainty in the technical analysis of a standard to mean that a rigid numerical outcome is inappropriate (note that while one option could be not to develop a standard if uncertainty is high, this might not be politically acceptable). 

Scientific Uncertainty and Science-Policy Interactions in the Risk Assessment of Hazardous Chemicals... 

Abstract In this chapter relatively recent European Commission risk assessment reports for three potential PBT/vPvB chemicals are used as examples to illustrate scientific uncertainty in the risk assessment process, and how science and policy interact when such uncertainty is handled. The studied risk assessment reports are for pentabromodiphenylether (Penta), octabromodiphenylether (Octa), and decabromodiphenylether (Deca) and the analyses focus on the scientific basis for assessing the risk of potential PBT and vPvB properties as described in these documents. The purpose of this effort is to contribute to a discussion aiming at clarifying the nature of science-policy interactions, and improving the transparency of the risk assessment process. 

The European Commission risk assessment reports for the diphenylethers Penta (European Commission 2001), Octa (European Commission 2003b), and Deca (European Commission 2002 and 2004) were used to identify examples of scientific uncertainties in the risk assessment process for potential PBT/vPvB substances. A systematic search for indicators of scientific uncertainty in these documents was performed, including how these uncertainties are described and handled in the risk characterization and in the conclusions section, and taking into consideration the use of and weight given to non-standard data in the final conclusions. 

---