How to identify Out-of-Trend Results in Stability Studies?

An article in PharmTech from June 2013 (by Trajkovic-Jolevska et. al) deals with the methods to identify Out-of-Trend (OOT) results in ongoing stability studies.

With regard to stability studies, it is important to make the difference between Out-of-Specification (OOS) and Out-of-Trend (OOT). Both the pharmaceutical industry and authorities often misuse these two terms.

The article defines OOT results as those results which don't follow the expected trend, either in comparison with other stability batches or compared to previous results collected during a stability study. OOT results aren't necessarily OOS, but they don't look like a typical data point.
 Although OOT results are a serious problem, neither the scientific literature nor regulatory guidelines fully address them.

The aim of the study described in this Pharmtech article by Trajkovic-Jolevska et. al was to perform a statistical evaluation of the statistical methods used in the identification of OOS results.

The authors of this article present 4 statistical methods for the identification of OOT results:

• The regression control-chart-method
• The by-time-point method
• The slope-control-chart method
• The z-score method

The conclusion of this article is the statement that there is a tremendous need for a regulatory Guideline on OOT results in stability studies. This is exactly the next objective of ECA's QC working group: an SOP on the handling of OOT results. On 22-23 October 2014, the OOT Forum in Prague will present the Draft of ECA's SOP on OOT results.

The complete article can be found with the title "Methods for Identifying Out-of-Trend Results in Ongoing Stability Data".