|
The Barr Judgement of 1993 was the central starting point for the ensuing
discussion within the pharmaceutical industry about the GMP-compliant
handling of out-of-specification results.
It is the main objective of an OOS investigation to
find out the cause for a deviating analytical result (Was the mistake made
during production or in the lab?). At the same time, the procedures set up in
the laboratory must withstand the accusation of "testing into
compliance", i.e. testing must not simply be repeated until a
compliant result occurs. The accusation of "testing into
compliance" was one of the main points of criticism in the Barr
Judgement. The Barr Judgement - also known as Wolin Judgement - is a very
comprehensive court decision that gives detailed answers to many
analytical questions. One of them is: How many analytical tests are
necessary in order to disprove an OOS result? Even if the Barr Judgement
is not quite up to date any more, it still represents a good source of
information for OOS questions.
All those who would like to get an insight into the
most important lab-relevant statements of the Barr Judgement can find a
very interesting summary when clicking here.
Although many years have passed since the Barr
Judgement, OOS is still a topical issue as can be read in a Gold Sheet
article titled "OOS Investigations Remain Key FDA Target" in
this year's March issue.
Concrete examples can be found in FDA Warning Letters:
"Confirmed OOS results for blend uniformity were obtained and no
failure investigation have been performed. …. Additionally, there was no
evaluation of the impact of these failures on other batches of this
product currently on the market."
"Inadequate method validation in that OOS findings were discarded
without investigation the cause of the OOS results and analytical data was
selectively reported."
According to the March issue of the Gold Sheet, one
of the main points in the references to failure investigation in Warning
Letters is that the scope of failure investigation is not wide enough.
Today the Agency expects e.g. that, when investigating process or product
deviations, failure investigation is also performed on other batches in
order to find out whether other batches of the same product are also
affected.
|
