GMP OOs – the procedure


he OOS procedure (out of specification), is always unexpected and obviously, no one likes it. Usually it is a signal that something is not in order somewhere in the entire GMP production/GMP analysis system. In concrete terms, an OOS event also means that if you come across something like this in your organization, there are a few things you need to consider:

  • Solving the actual deviation from the specification, the OOS,
  • How robust is my GMP production process?
  • How robust are my QC sampling and QC testing procedures?
  • How well have I performed my process validation or cleaning validation?

That last point is even more relevant when it’s been a long time since you’ve completed your validation work. Or worse, when there’s doubts about the quality of the validation because back then, people were a bit busy with getting the factory up and running.

The OOS procedure

Besides the annoying fact that you have an OOS, you can use the situation at the same time to improve your overall quality level. Take the OOS procedure for example. The GMP OOS procedure should be specific. It follows a path similar to that of a deviation, but (for historical reasons) an OOS requires a more specific approach.

The MHRA and US-FDA (amongst others) have written excellent guidelines / documents, which can be converted into company-specific OOS procedures. Whether you use these documents as input is up to you. But always keep in mind that an OOS specifically concerns the release tests and that these are designated as such in the registration files.

For this reason, it is industry practice for pharmaceutical companies to use the term “specification” only for those specific tests. Especially to avoid confusion among personnel and inspectorates/GMP auditors.


There are two specific cases that cannot be handled with a standard OOS procedure, these are sterility failures and media fill failures. Nor can these be handled with a standard deviation procedure. For that reason separate GMP SOP’s must be drawn up for these two cases.

Another “exceptional” type of OOS: the ongoing stability OOS, deserves special attention also. After all, an OOS on your ongoing stability could be an indication that something is wrong with the product itself (or at least the product stability).

quality culture

It is important that the quality culture of a company is such that you always find the root-cause of the OOS. In particular, you must conduct a thorough investigation into what exactly is the (most likely) cause of the OOS. But while you are at it, it’s important to assess all other possible causes as well. After all, those less likely causes could potentially cause problems in the future.

Note that governments have special GMP legislation for reporting any OOS. It’s important to read up on those guidelines. Especially when it comes to medicinal products which are prone to cause shortages when they are discarded as a result of an OOS.

Something we see quite often is that the search for (the most likely) cause is aborted too quickly. Usually because it costs too much money. The batch is subsequently destroyed. This should not occur in a quality culture, not so much the destruction of the batch but much more the discontinuation of the investigation after this decision.


Last but not least, it is important to realize that you shouldn’t keep going endlessly to find the exact, super-root-cause of an OOS. It’s nearly impossible to do and makes little sense. Find the most likely root-cause to which you, your team, the SME’s and management agree with. And as mentioned above: do not forget the assessment of less possible causes. The latter category can in turn yield quality / business improvements!

About the author

Add Comment