SOP for deviations in QC

[Irene59]

CAP inspection is looming, and dept director asked if there was a procedure in place for this.

Is there a specific blanket SOP that I can write for all the "tube tests" that would cover everything from ABO/Rh to antigen screening, for QC deviation? I don't just mean a blanket "document results on XYZ quality improvement form." I am looking for something to add to each procedure, or a separate procedure entirely that states what exactly to do, i.e., check expiration date of reagent, re-test reagent, etc. etc.

For example, in chemistry, the procedure would look like: if QC is outside of acceptable limits, recalibrate, reconstitute, rerun, etc. etc. I have not seen anything like this anywhere for BB

[David Saikin]

What would you do if there was a "deviation" in you bb qc? That's what you write in your procedure . . . rerun or/then open a new vial. There are not deviations in bb qc - it either works or doesn't (we don't have 2/3 standard deviation qc algorhythms). Check your reagents/check your qc kit/system. I've never seen bb qc not work.

[KKidd]

For all reagents that are included in your daily QC, you could include a statement in that procedure. For my antigen typing, the facility procedure refers to the product insert for the specific antigen (E, K,etc.)

I agree with Davd! BB QC doesn't usually fail. Tha last problem I had was with my cellwasher, not my reagents. That was a long time ago. Makes me glad our testing is not like Chemistry or Hemo.

:highfive:

[clmergen]

We wrote into our QC procedure, if it fails once - rerun the test, if it fails again - open a new bottle, if it fails again - open a new lot number and notify your supervisor.

[Malcolm Needs ☆]

If it still fails - resign!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

:eek::eek:

[RR1]

The main points with response so far is that you are not meeting regulatory requirements for handling out of specification results (QC failures). If you simply re-run the control when you are unsure of what the reason for failure was this is known as "testing into compliance" which is not allowed.

[adiescast]

The main points with response so far is that you are not meeting regulatory requirements for handling out of specification results (QC failures). If you simply re-run the control when you are unsure of what the reason for failure was this is known as "testing into compliance" which is not allowed.

What is your policy for this, Rashmi?

[RR1]

What is your policy for this, Rashmi?

Hi adiescast, in the UK section: Quality area, Irene59 has asked the same question and I have attached the SOP that we use for Out-of-spec testing.

OOS testing is not just about serology tests- but any (validation, computer checks etc) where the main point is that an unexpected result is checked to explain why this happened and if you can't explain the anomaly the repeat must be performed fairly and show the "correct" result is consistently repeatable.

The MHRA cited us for "testing into compliance" during one of our inspections, where we had recorded a failed QC that was re-run and passed first time. On the internet the policy for OOS testing seems to be a fairly hot topic with the FDA too.

[BBK710]

We wrote into our QC procedure, if it fails once - rerun the test, if it fails again - open a new bottle, if it fails again - open a new lot number and notify your supervisor.

We do the same. I have never seen the QC fail. I think if it did it would probably be due to contamination which would be proved by repeating and/or opening a new bottle of reagent. I don't think that this is testing it into compliance but determining what the problem is with the QC and resolving the problelm to assure accurate patient testing.

[mcgouc]

I totally agree that our BB reagent QC doesn't usually fail, but we have a policy to follow if there is a problem. Part of our daily QC is to document the lot number, expiration date, and visual checks. If reactions are not acceptable, we document the original reactions and repeat the testing. Usually, this resolves the problem - the tech remembers he forgot to add something and this is documented on form. If this doesn't resolve the problem, our policy is to test a new vial of same lot number and a vial of a different lot number (if these are available). If the product does appear to be "defective", we quarantine and attempt to determine reason - is one bottle contaminated or whole lot number not reacting correctly? If we contaminated a bottle, what did we do to contaminate it? If we cannot resolve the problem, we notify reagent company. All control failures and corrective actions are documented and trended on a spreadsheet. Looking back over the past 4 years, I don't find any times where the problem with daily QC wasn't tech error. However, during the same period, we have notified manufacturers regarding problems with screen cells and fetal screen kits that were not detected when we performed reagent QC, but were found in routine use.

[RR1]

Something else to think about is that if you can't resolve the problem how do you address all testing that was performed PRIOR to the failed QC and up to the last good QC?

The QC tests all reagents used for the test itself and the operators performing the test; if one of your staff submitted an incorrect results on an external QC (NEQAS, or CLIA) ,would you not also want to check their performance and also audit some of their previous work to see if any patient results had been compromised too?

[clmergen]

We have found more problems with reagents during testing than at QC testing time. I think with the forward and reverse type, you have built in checks for the ABO, we have check cells for the screening reagents. I don't see it as an issue to be concerned about.