Benchmark Dose Analysis via Nonparametric Regression Modeling

• Published in: Risk analysis Vol. 34; no. 1; pp. 135 - 151
• Main Authors: Piegorsch, Walter W., Xiong, Hui, Bhattacharya, Rabi N., Lin, Lizhen
• Format: Journal Article
• Language: English
• Published: Hoboken, NJ Blackwell Publishing Ltd 01.01.2014; Wiley
• Subjects: Animals; Benchmark analysis; Benchmarking; Benchmarking - statistics & numerical data; Benchmarks; Biological and medical sciences; BMD; BMDL; bootstrap confidence limits; Bootstrap mechanism; Bootstrap method; Carcinogens - toxicity; Computer Simulation; Confidence limits; dose-response analysis; Dose-Response Relationship, Drug; Evaluation; Formaldehyde - toxicity; General aspects. Methods; Humans; isotonic regression; Mathematical models; Medical sciences; Models, Statistical; Monte Carlo Method; Regression; Regression Analysis; Risk; Risk analysis; Risk assessment; Risk Assessment - methods; Risk Assessment - statistics & numerical data; Samples; Simulation techniques; Statistics, Nonparametric; Studies; Toxicological Phenomena; toxicological risk assessment; Toxicology; Uncertainty; bootstrap confidence limits; isotonic regression; Toxicity; Confidence limit; Regression; dose-response analysis; Risk analysis; BMDL; Modeling; toxicological risk assessment; Dose activity relation; Toxicology; Risk evaluation; Benchmark calculation; Risk assessment; BMD; Bootstrap; Benchmark analysis
• ISSN: 0272-4332; 1539-6924
• DOI: 10.1111/risa.12066

Estimation of benchmark doses (BMDs) in quantitative risk assessment traditionally is based upon parametric dose‐response modeling. It is a well‐known concern, however, that if the chosen parametric model is uncertain and/or misspecified, inaccurate and possibly unsafe low‐dose inferences can result. We describe a nonparametric approach for estimating BMDs with quantal‐response data based on an isotonic regression method, and also study use of corresponding, nonparametric, bootstrap‐based confidence limits for the BMD. We explore the confidence limits’ small‐sample properties via a simulation study, and illustrate the calculations with an example from cancer risk assessment. It is seen that this nonparametric approach can provide a useful alternative for BMD estimation when faced with the problem of parametric model uncertainty.