A firm reminder to the general manager of Dongying Tiandong Pharmaceutical Co., Ltd. in Shandong, China from the Food and Drug Administration:
1. Failure to adequately investigate and document out-of-specification results according to a procedure, and implement appropriate corrective actions.
...For example, according to your Deviation Handling Sheet No.07-2015021, you resampled and tested crude heparin batch Y102-1504005 multiple times, with the following results.
You neither evaluated the initial sample OOS, nor conducted retesting of the initial original sample to confirm it. Instead, you resampled until you obtained a passing result.I love that last sentence.
Similarly, your initial test results for another crude heparin batch (Y102-1503008) were also OOS. Again, you resampled without justification, and accepted the batch when you obtained results within specification.
Disregarding the OOS results, and resampling and retesting without scientific justification, constitutes “testing into compliance.” This practice is unscientific and objectionable under CGMP....
I've seen this in every industry job I've had (all both of them!), and although my critical mind is quick to denounce such practices, there is good reason things like this happen. If you have a test that has enough false positives you'll start to see every positive as false. Good reminder post, CJ.
ReplyDeleteThe casual "oh, there are a lot of false positives on that test" can mean a couple of different things:
Delete1) "We don't always get the result we want"
2) "We didn't spend much time characterizing and qualifying the assay before we wrote the SOP"
3) "We didn't really think about alternate assays, or compare them"
4) "This assay has a reproducible output of irreproducible results"
In my experience, there are a lot of 1s, 2s and 3s and not so many 4s.
Hang in there, Dongying Tiandong Pharmaceutical Co. In 2017 the new administration will begin to eliminate those productivity-killing regulations.
ReplyDeletestill waiting....
Delete