I’m happy to have the chance to update the site with another editorial, and amazingly, it still takes little time to think of another topic to pencil in for posting. There are so many to choose from. Finding the time to write them can be another story, but here I am.
I’m on yet another ‘crossing’ tonight, and I have a chance to catch my breath, relax, and write another piece. After pushing back from the gate, taxiing to the runway and rotating, we’re finally on the way to our destination, sans raison d’être from the cockpit avionics resulting in stationary status. After yet another lengthy delay, I happily accept the turbulence, and here’s hoping that it won’t hinder my typing too much. Nonetheless, I’m delighted to be on my way.
What an undertaking it can be to experience the miracle of flight without incidence. I’ve not had the best of luck in my travels lately. For that matter, what an undertaking it can be to oversee the initial and ongoing qualification of multiple contract calibration labs.
The FDA and other regulatory bodies require us to qualify contract calibration labs (as well as other contractors performing work on our regulatory assets), and it can be very taxing to verify and subsequently document (if it’s not documented, it didn’t happen) the quality programs of the multiple vendors that so many of us use. If a contractor fails to meet GxP requirements for the work performed on their behalf, we the customers are ultimately held reliable for failing to recognize and/or failing to apply corrective action to the substandard work.
Just as our metrology programs are audited on a cyclical basis to maintain and verify quality, it is just as necessary for us to periodically re-verify our contract labs quality programs after an initial assessment.
Aside from verifying that a CCL (contract calibration laboratory) is technically capable of performing the applicable calibration discipline to our standards, there are a number of additional quality related items that should be assessed and documented, or at the very least, strongly considered:
- Existence of a detailed quality manual, written to an accepted standard
- Personnel have the necessary training, education, experience, or any combination thereof, to perform the applicable disciplines within the organization
- There are generally accepted procedures documenting the calibration discipline of concern
- The laboratory uses standards that are directly traceable (in an unbroken chain) to SI, and there is a documented cumulative uncertainty budget all the way back to the applicable SI Unit (International System of Units – The worldwide standard for measurement units)
- There is a well-defined time period where the contractor will maintain copies of the certificates generated for the calibrations performed for each asset, as well as any related supporting documentation
- There are defined, written procedures detailing the approach to reverse traceability and customer notifications in the event of standard recalls internal to the laboratory
- The laboratory follows a recognized metrological standard for the applicable calibration disciplines to be performed
- The laboratory has the infrastructure to maintain the environmental conditions necessary to support the calibration and stated uncertainty
- The laboratory can generally find it’s way out of a paper bag
Customer service assessments are not out of the scope of an assessment either.
The best way, hands-down, to insure that the CCL is qualified to meet your requirements is to simply schedule periodic onsite audits of a contract calibration laboratory. Since many of us lack the time or resources to drop in on each and every contract lab we use, that option is often times not within reality.
As many of you have no doubt experienced, asking to remotely review a contract laboratory’s quality manual often poses its own unique challenges. Many cal labs consider these documents to be proprietary, and will only share them with an auditor while onsite. Now for the irony – A complete quality manual is a key prerequisite before a company can feasibly be hired to provide services in many regulated industries. If you can’t review the company’s quality manual remotely and can’t spare the resources for an onsite audit, what’s left?
Let’s consider the above as ‘Plan B’. Fortunately there is an easier approach to all of this, taking advantage of a system that is funded each year by many labs already. I will refer to this as ‘Plan A’.
In short, it’s nice to let others do our job for us.
ISO has a very well defined standard for metrology called ISO/IEC 17025 (formerly Guide 25). This standard, for the most part, greatly exceeds and encompasses the requirements of many regulatory bodies for calibration, and is considered the world standard for quality metrology. There are entities referred to as ‘Accreditation Bodies’ (AB’s) that are themselves ”accredited to perform accreditations” on calibration labs to the ISO 17025 standard. Typically, AB’s that are assessed and signatories of a mutual recognition agreement with ILAC (International Laboratory Accreditation Cooperation) are considered acceptable on a world standard. There are now three of these accreditation bodies within the USA that are signatories, which are ultimately ”licensed to accredit” to ISO 17025 per ILAC standards.
When a laboratory is accredited and successfully demonstrates competency, it receives a certificate of accreditation from the AB for the calibration discipline assessed, alongside a stated uncertainty budget for that discipline. Those certificates are readily available for public view on the AB’s website. Interestingly, a good number of contract labs fail the first visit(s) from an AB. It’s not an easy task to prepare for and prove that you’re competent in ISO 17025 audit. Once successful, the scope of accreditation is valid for one year.
If you’ve not written an SOP dictating your approach to qualifying contract calibration labs, I would suggest doing so. It should include specific criteria that details what necessary quality programs should be in place for a prospective lab to get the ‘green light’. It should also detail the documentation of audits for labs that make the grade. Keeping a well-updated file for each lab hired is important.
My SOP had a Plan A / Plan B approach to all of this. Plan ‘A’ stated that if a lab was accredited to ISO 17025 for the calibration discipline required, and had an acceptable uncertainty budget for the asset used in the field, printing out a copy of the certificate of accreditation and filing it was all that was needed. We kept a close eye on the upcoming dates for re-accreditation, and updated the files when the labs got a new certificate each year.
Plan ‘B’ was much more cumbersome, because we needed to document the items I listed above, in even more detail.
Now, keep in mind a few things. First of all, ISO 17025 accreditation is not a ‘cure all’ for quality. There are holes in any quality system – none of us are perfect. It does, however, demonstrate to a reasonable extent the laboratories competency. In addition, I would consider documenting precisely what the 17025 requirements are in the SOP for qualifying contract labs, so there is little, if any, room for questions by an auditor that these requirements encompass those of the regulatory bodies that oversee us.
You might be interested to know that much of the driving force for accreditation by contract labs in the US came from the greatly varied requirements of their customers. FAA regulated customers had one requirement. DOE might have had another. FDA customers had their own stipulations. ISO might have had yet another requirement. Each one of these regulated industries, while having their own requirements, ultimately respect the documented traceability chain back to the SI unit, and accreditation is one of the avenues to remove this smokescreen.
We have begun to take the road to global harmonization. In Europe, contracting to a lab without accreditation to 17025 is unthinkable. For that matter, I have heard some interesting rumors that the FDA is taking a closer look at 17025, and it doesn’t surprise me. Just as QSIT (Quality Systems Inspection Technique) was driven by scrutiny from global standards, I would assume that they would not overlook a well-defined metrology standard either.