A new approach for dealing with the stability of drugs in biological fluids.

The purpose of this study was to develop a procedure for investigating the stability of drugs in biological fluids based on sound experimental design and to use a statistical procedure which would allow conclusions to be made concerning stability with an acceptable degree of certainty (95%). The experimental procedure involved replicate analysis of the drug in stored and freshly prepared samples on the same day. The relative difference in response between these two sets of samples and a 90% confidence interval for the true change in response was calculated. This confidence interval enabled us to detect a pharmacokinetically relevant degradation. It is argued that this approach is superior to the stability test procedures based on the t test. The application of the method to the assessment of the stability of three different drugs in biological fluids is described.