Estimating the maximum effective dose in a quantitative dose-response experiment.

Abstract

A simulation study was conducted to compare several procedures for estimating the maximum effective dose in a quantitative dose-response experiment. Using four equally spaced dose levels, data were generated from four different model types: the quadratic growth curve, the Mitcherlich growth curve, the linear-linear plateau spline model, and the quadratic-linear plateau spline model. Each model type was parameterized to create three different model ranges, and for each range, data were generated from populations with three different standard deviations. The existence of unique dose-response curves is assumed; thus, all the procedures compared in this paper require that the data have been modeled by a polynomial or nonlinear regression model. An attempt was made to fit each generated data set with each of the four model types. Maximum effective dose estimation procedures were applied to a data set only when the data were adequately described by a given model. The stimulation indicated that the estimate of the maximum effective dose is influenced more by the choice of model than by the method of estimation. Because of the consistently low estimates produced when the data were modeled by the linear-linear plateau spline, this model is not recommended for use an maximum effective dose estimation experiments. The simulation also demonstrated that the design failed to provide sufficient information about the form of the dose-response curve. Designs with more than four dose levels should be considered.