Using sparse dose–response data for wildlife risk assessment

Abstract

Hazard quotients based on a point-estimate comparison of exposure to a toxicity reference value (TRV) are commonly used to characterize risks for wildlife. Quotients may be appropriate for screening-level assessments but should be avoided in detailed assessments, because they provide little insight regarding the likely magnitude of effects and associated uncertainty. To better characterize risks to wildlife and support more informed decision making, practitioners should make full use of available dose-response data. First, relevant studies should be compiled and data extracted. Data extractions are not trivialpractitioners must evaluate the potential use of each study or its components, extract numerous variables, and in some cases, calculate variables of interest. Second, plots should be used to thoroughly explore the data, especially in the range of doses relevant to a given risk assessment. Plots should be used to understand variation in dose-response among studies, species, and other factors. Finally, quantitative dose-response models should be considered if they are likely to provide an improved basis for decision making. The most common dose-response models are simple models for data from a particular study for a particular species, using generalized linear models or other models appropriate for a given endpoint. Although simple models work well in some instances, they generally do not reflect the full breadth of information in a dose-response data set, because they apply only for particular studies, species, and endpoints. More advanced models are available that explicitly account for variation among studies and species, or that standardize multiple endpoints to a common response variable. Application of these models may be useful in some cases when data are abundant, but there are challenges to implementing and interpreting such models when data are sparse. Integr Environ Assess Manag 2014;10:3-11. © 2013 SETAC

Benchmark dose analysis framework for developing wildlife toxicity reference values

“…Commercial statistical software packages (e.g., SAS, MatLab, GraphPad, SigmaPlot) contain options for data analysis using linear or nonlinear regression, or other specific dose–response models. Risk assessment scientists have also published graphical approaches, spreadsheet‐based tools, or independently developed computer codes (Caux and Moore ; Hill et al ; Ritz et al ). Similarly, environmental agencies have developed software tools and guidance documents to conduct dose–response analysis for regulatory purposes, and at present these are freely available for public use.…”

“…Finally, it can be useful to perform exploratory analysis of the raw data (prior to modeling) to identify dose–response trends, outliers, or uncertain data sets. Hill et al () provide a general method for data visualization by plotting the normalized effect responses to control responses (i.e., response of treatment group divided by response of control group), which allows for comparison of different data sets in a consistent fashion. The aforementioned methods provide a structured framework for data collection and preparation to aid the assembly of quantitative information for TRV development.…”

Methylmercury Effects on Birds: A Review, Meta‐Analysis, and Development of Toxicity Reference Values for Injury Assessment Based on Tissue Residues and Diet

“…Thus, to consolidate these four types of endpoint responses such that the effect of Hg on birds was always expressed as a negative relationship and bounded between % CNR = 100% (no effect) and %CNR = 0% (full effect), raw data transformations were necessary for the positive relationships (types 2 and 4 just listed) but not for negative relationships (types 1 and 3 just listed; see Table 1). Similar to Dillon et al (2010) and Hill et al (2014), we transformed proportion data (y) with a positive relationship (type 2 above) using a − y 1 transformation (Table 1) to get the converse of the data that were originally reported (e.g., 60% mortality became 40% survival). For nonproportion data with a positive relationship (type 4 above), the treatment-to-control response ratio generally ranged from 1 (when treatment and control responses were similar) to higher values (as treatment responses exceeded control responses); thus we used a /y 1 transformation to convert the response into a value with a [ ] 0,1 range and negative relationship with Hg (Table 1).…”

“…This type of transformation was not used in Dillon et al (2010), and they did not include endpoints of nonproportion data with positive relationships in their assessment. Hill et al (2014) recognized the difficulty of including data from positive relationships with nonproportion data (type 4 above), and stated that "care should be taken in normalizing such endpoints to control performance." Similar to the use of − y 1 to invert positive relationships in the [ ] 0,1 range into negative relationships in the [ ] 0,1 range, the /y 1 transformation is a natural choice for inverting positive relationships in the [ ∞] 1, range into negative relationships in the [ ] 0,1 range.…”

Wildlife ecological risk assessment in the 21st century: Promising technologies to assess toxicological effects

“…For wildlife ERAs, DR relationships are recommended over no‐observed‐adverse‐effect level (NOAEL) and lowest‐observed‐adverse‐effect level (LOAEL) values, in part because they provide information about the magnitude and severity of responses if an effect threshold is exceeded, and they can identify thresholds of interest (e.g., 20% effect concentration, EC20) instead of relying solely on the statistical power of hypothesis tests (Allard et al, 2010; Mayfield et al, 2014). Hill et al (2014) provided a useful review of approaches to DR analysis, and additional discussions on model selection and fit, and software options are available (e.g., Erickson & Rattner, 2020; Mayfield & Skall, 2018). Nonstandard, lower‐tier endpoints can be directly or indirectly modeled by incorporating effects into a DR model of a higher‐tier endpoint.…”