In vitro-in vivo correlation, a time scaling problem? Evaluation of mean times.

A new algorithm for the calculation of the mean time and further statistical moments is presented by means of a mathematical deduction which is outlined in brief. The graphical interpretation of the mathematical results demonstrates that the calculation is easy to perform. Furthermore the method is illustrated in detail with data from a sustained release formulation of theophylline. These data were obtained from a clinical study and from an in vitro dissolution test with the same formulation. Statistical analysis revealed that the rate of release in vivo was slower than in vitro by a factor of 0.7, using a commercial dissolution model. Implications of this approach with respect to in vitro - in vivo correlations are discussed.