Interchangeability and predictive performance of empirical tolerance models

Models of tolerance are commonly derived on empirical grounds, because of l ack of knowledge about the mechanism of tolerance or because of the difficu lty of appropriately simplifying complex physiological processes. The prese nt study was performed to evaluate the interchangeability of tolerance mode ls used in the literature and to address some determinants for selection of an appropriate design and data analysis strategy. Seven models were chosen (noncompetitive antagonist model, partial agonist model, reverse agonist model, direct moderator model, indirect moderator mo del, pool model and adaptive pool model) along with their corresponding par ameter estimates, representing a wide range of empirical models. The performance of the models on various data sets was evaluated. Data were simulated from each original model and were further analysed by the other models. The effect-time course of each and every data set could be describe d well by at least 2 different empirical tolerance models, but no model cou ld describe all the data sets adequately. However, all models could adequat ely describe at least 2 different data sets. This indicates that, without a dditional knowledge or assumptions. it is unlikely that reliable mechanisti c information can be deduced from the mere fact that 1 (or more) of these m odels can describe the data. Generally, data expressing only limited tolera nce can be described by a wide variety of models. whereas few models will b e appropriate for data characterised hy extensive tolerance. The models that pave an adequate description of a data set were selected fo r further study that investigated their predictive capacity based on the pa rameters previously determined. Predictions were made for 4 different admin istration schemes. The selected models gave similar predictions for the ext ended designs of 3 data sets for which the original study designs character ised tolerance well. For the other 4 data sets, the selected models gave di sparate predictions, although the models described the original data set we ll. Thus, the predictive capability of a model was linked to the original s tudy design, whereas the correlation between predictive performance and the type of model was weak or absent. Based on the results. factors of importa nce for the design and evaluation of studies of tolerance were identified a nd discussed.