Out-of-Specification Micro Result Investigation
Scenario brief
A finished product batch has failed its Enterobacteriaceae limit. Lead a phase I lab investigation and decide whether the result is valid.
- About 6 min
- 6 parts
What you'll do
This is the procedure you'll follow at the bench. Open a part to read it step by step, with the reason for each step, before you start.
- 1
Gown up5 steps, in order
Choose the right protection for the work and put it on in the right order.
The method for this part
- 1
Tie hair back, remove jewellery
First, before your hands are washed: you won't touch your head again once they're clean.
- 2
Lab coat, fastened
The coat goes on before you wash, so clean hands don't have to handle it, and before gloves, so the gloves can cover its cuffs.
- 3
Wash your hands
Hands are washed once your hair and the coat are dealt with, and before you touch clean eye protection and gloves. With rings and watch off, the whole hand and wrist can be reached.
- 4
Safety glasses
Eye protection goes on before gloves, with freshly washed hands.
- 5
Nitrile gloves
Gloves go on last, over the cuffs and onto clean, dry hands, so they're clean when you start work.
What goes wrong here
- Respirator. Routine bench work doesn't need a respirator. Wearing more than the risk calls for just makes the work clumsier.
- Plastic kitchen apron. A kitchen apron leaves your arms bare. In a lab you need a fastened coat with sleeves.
- Your own jacket. Outdoor clothes stay in the locker: they carry contamination in and out.
- 1
- 2
Collect your kit3 things to fetch
Fetch what the test needs from the stores around the lab, or have it brought to you.
The method for this part
What you check
- • Pen
- • Investigation record
- • Laboratory records
What you decide by
- • Take only what the test uses: anything extra on the bench is clutter to reach over.
- 3
Secure and report7 steps
Protect the evidence and tell the right people, before anything is repeated.
The method for this part
- 1
Stop, and keep everything from the test
Stop work on this batch. Leave the plates from the failed test, the dilutions, the sample and the worksheet exactly as they are.
Everything from the test is evidence, and evidence thrown away can never be checked.
- 2
Secure the plates in the fridge
Label the plates 'under investigation' and put them in the sample fridge.
Kept cold, the plates can be read again and the colonies identified.
- 3
Secure what is left of the sample
Seal the retained sample, label it and put it in the sample fridge.
Any repeat the procedure allows has to come from the same sample, properly kept.
- 4
Tell the laboratory supervisor
Report the result to your supervisor now, before you try to explain it.
An out-of-specification result is reported at once, not once it has been tidied up.
- 5
Ask the quality manager to hold the batch
Tell the quality manager, so that the batch is put on hold.
A batch under investigation must not be able to leave the site.
- 6
Open an investigation record
Open the investigation record: write down the result, the time, who was told and what has been secured.
An investigation that isn't written down as it happens can't be relied on afterwards.
- 7
Pull the records for that run
Pull the laboratory records for that run: media, equipment, controls, training and monitoring. Do it before anything is repeated.
A repeat is only justified by a documented laboratory cause, so the evidence comes first.
What goes wrong here
- Retest the sample straight away. A retest before any cause is shown is retesting into compliance: a pass would leave the failing result unexplained.
- Discard the plates. The plates are the evidence: once binned, they can't be read again or the colonies identified.
- Average the fail with a retest. An average hides a failing result inside a number that looks acceptable.
- 1
- 4
Review the evidence8 records to judge
Judge what each record shows about the laboratory's part in the result.
The method for this part
What you check
- • Read the record
What you decide by
- • Ruled out: the record is complete and shows this part of the work was in control.
- • Can't be ruled out: a record is missing or a check is overdue, so nothing is shown either way.
- • Assignable cause: a documented fault that could have produced the high count.
- • For this exercise the limit is 100 CFU/g, and the method is a pour plate incubated at 37 °C for 24 hours.
What you can decide
Ruled out · Can't be ruled out · Assignable cause
- 5
Conclude the investigation4 questions
Decide what the evidence allows.
- 6
De-gown4 steps, in order
Take it off in the right order, most contaminated first, and wash your hands last.
The method for this part
- 1
Take off: nitrile gloves
Gloves are the most contaminated thing you're wearing: they come off first, peeled inside out.
- 2
Take off: safety glasses
Glasses come off by the arms once the gloves are off, so you don't touch your face with dirty gloves.
- 3
Take off: lab coat, fastened
The coat comes off after the gloves and glasses, and stays in the lab on its hook.
- 4
Wash your hands
Hands are washed last, after everything that might have contaminated them is off, and before you touch the door.
- 1
What you'll take away
- Stop, keep everything and report at once: the plates, the sample and the worksheet are evidence.
- The batch goes on hold, and the investigation is written down from the first minute.
- Review the evidence before anything is repeated: a retest is only justified by a documented laboratory cause.
- A gap in the records is something to put right, not a cause. Only documented evidence of an error invalidates a result.
- With no assignable laboratory cause the result stands: the batch stays on hold and the investigation moves to manufacturing.
- Never retest until it passes, average a fail away, or discard the evidence.
Equipment unlocked
- Investigation log
- Media QC records