New generalized poisson mixture model for bimodal count data with drug effect: An application to rodent brief‐access taste aversion experiments

Abstract

Pharmacodynamic (PD) count data can exhibit bimodality and nonequidispersion complicating the inclusion of drug effect. The purpose of this study was to explore four different mixture distribution models for bimodal count data by including both drug effect and distribution truncation. An example dataset, which exhibited bimodal pattern, was from rodent brief‐access taste aversion (BATA) experiments to assess the bitterness of ascending concentrations of an aversive tasting drug. The two generalized Poisson mixture models performed the best and was flexible to explain both under and overdispersion. A sigmoid maximum effect (Emax) model with logistic transformation was introduced to link the drug effect to the data partition within each distribution. Predicted density‐histogram plot is suggested as a model evaluation tool due to its capability to directly compare the model predicted density with the histogram from raw data. The modeling approach presented here could form a useful strategy for modeling similar count data types.

Taste evaluation of a novel midazolam tablet for pediatric patients: In vitro drug dissolution, in vivo animal taste aversion and clinical taste perception profiles

“…Rat aversion data were analyzed using the R statistical software (http://www.Rproject.org) after removal of data sets with  1 lick(19), and the comparison of means was performed using 2-tailed unpaired t-test. Non-parametric Mann-Whitney test and 2-tailed unpaired t-test were applied on the clinical taste evaluation data.…”

Modelling Bimodal Data Using a Multivariate Triangular-Linked Distribution

“…There is a vast number of distributions to choose from, thus when it comes to creating a mixture model, there are virtually endless possibilities and the final decision will typically depend on the data and the needs of the practitioner. Sheng et al [9] demonstrated that mixture models can be used in pharmacodynamic studies in which bimodal count data arise. They found that the two generalized Poisson mixture model was the best fit for their bimodal dataset consisting of the number of times that a rodent licked an oral medication in a palatability study.…”