Some considerations about stability study design

The aim of a stability study is to check whether or not a drug maintains its initial properties (over a given range) in a period of time. There are many parameters considered in a stability study, but in this paper only the active content will be considered. Evaluation of this parameter is performed by measuring the content of the active ingredient in the drug at different times and on expiration to see if it is within specification (1). The problem also involves estimating whether the drug will perform satisfactorily and maintain its initial properties in the future. In this paper, we provide a criterion for the evaluation of experimental design used in a stability study for predicting the content of an active ingredient in a drug, starting from the confidence limit calculated following International Conference on Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use (ICH) recommendations. The point of view adopted is akin to hypothesis testing. The main question is this: "If the drug fulfills the requirements at a given time in the future, what is the probability that the data collected for that drug will show its suitability?" For this purpose, two concepts are used-producer gain and consumer loss-which are defined as follows: producer gain is the advantage (for the producer) of recognizing a drug as "good" (provided it is good and within specification); consumer loss is the loss (for the consumer) when the drug is no longer good, but the data collected indicate that it is good. The aim of the experimental design in this study is to increase the probability of producer gain and maximize it on the basis of a given consumer loss.