Quantifying detrimental risks from hazardous inputs is a critical component in disease risk evaluation. Although statistical methods in risk analysis have been well-studied for a single hazardous exposure, a critical gap exists in the practical development of these methods when mixed input factors might affect the adverse outcome(s) under study. The focus of this small, self-contained research project is to develop and study new methods of statistical inference for use in quantitative risk assessment, with focus on mixed quantitative and qualitative input factors. Project emphasis will center on estimation of benchmark doses (BMDs) associated with a pre- selected level of benchmarked response (BMR). Motivated by a translational risk-analytic application in childhood respiratory disease, this research project will develop methods for developing benchmark markers (the ?doses?) of an adverse health outcome when additional, qualitative, multi-level factors are critical to the analysis. The theoretical, asymptotic characteristics of the estimators will be established, and from these large- sample benchmark dose confidence limits (BMDLs) will be constructed. Besides the targeted application to childhood respiratory health, translational application is anticipated for any biomedical, toxicological, pharmacological, environmental, or occupational studies where human or animal data are used to set benchmark markers or other safe, low-dose levels with mixed inputs. The resulting guidelines will improve public health planning and disease risk management when dealing with joint exposures where combinations of harmful agents/mixed inputs may interact or otherwise act jointly to amplify risk of disease. The research will fill critical gaps in multi-factor benchmark analysis, and will have potential translational application to non-standard/non-traditional settings across a wide variety of public health scenarios. Evaluation phases of the project will study the small-sample operating characteristics of the estimators via Monte Carlo computer calculations, and will explore the methodology's capabilities for practical application with motivating data on childhood asthma.
The aim will be to assess the methodology's (1) practical capabilities, and (2) its viability for estimating benchmark doses in these translational settings. The validated procedures for BMD estimation and for calculating the confidence limits will be coded for public use and ported to download site(s) on the Internet to facilitate dispersal across the widest possible corps of users.
/Relevance This study involves exploratory statistical research to develop benchmark or other safe exposure levels for multiple inputs that can predict an adverse health outcome. The methods expand upon traditional, single-input benchmark risk analyses and provide insight into the scientifically pressing issue of how to incorporate mixed, concomitant, multiple-level factors under the benchmark risk estimation paradigm. The work will develop multi- factor benchmark dose inferences, whose resulting guidelines can improve public health planning and disease management. Targeted application focuses on selected issues in childhood health, although translational applications are anticipated for any biomedical, environmental, toxicological, pharmacological, or occupational studies where human or animal data are used to set benchmark or other safe, low-dose levels of one or more hazardous stimuli or markers.
| Schissler, A Grant; Piegorsch, Walter W; Lussier, Yves A (2018) Testing for differentially expressed genetic pathways with single-subject N-of-1 data in the presence of inter-gene correlation. Stat Methods Med Res 27:3797-3813 |
| Liu, Jingyu; Piegorsch, Walter W; Schissler, A Grant et al. (2018) Autologistic models for benchmark risk or vulnerability assessment of urban terrorism outcomes. J R Stat Soc Ser A Stat Soc 181:803-823 |
| Wheeler, Matthew W; Piegorsch, Walter W; Bailer, Albert John (2018) Quantal Risk Assessment Database: A Database for Exploring Patterns in Quantal Dose-Response Data in Risk Assessment and its Application to Develop Priors for Bayesian Dose-Response Analysis. Risk Anal : |
| Piegorsch, Walter W (2017) Are p-values under attack? Contribution to the discussion of 'A critical evaluation of the current ""p-value controversy""?'. Biom J 59:889-891 |
