Posted by Sten Westgard, MS and James O. Westgard, PhD
As reported elsewhere on the website, several of the symposiums this year were on standardization and harmonization for various analytes. This year, however, there was an added note of urgency. For many laboratorians, standardization seems like a somewhat esoteric subject, important but not necessarily pressing. But with the expectation of widespread implementation of electronic medical records, the problems with “comparability” of test results (and analytical methods) are about to become stark.
If we achieve a single electronic medical record for a patient, what happens when all the different laboratory tests, from all the different settings (outpatient clinic, hospital, clinician office), from all the different methods (different manufacturer methods, POC vs. central lab instruments, etc.), get presented, even plotted, together on a single chart? Will all that non-standard noise look like a condition requiring new treatment? Or will the clinician be only further confused about the patient’s real status, requiring yet another round of testing?
If we still have bias between our methods, those differences, which used to be hidden because the results were never available to be compared, are about to reveal all the warts in our performance.
Unless we standardize, harmonize, and improve our methods, those biases are going to be passed on to the clinician, to confuse and consternate them.
Unfortunately, wanting standardization and harmonization is not enough. For many analytes, we need a system of reference methods and materials to establish the traceability of routine methods. While there are regulatory requirements for traceability in the European Union, there are none in the US. That gap in the US FDA/CLIA regulatory framework is now coming back to haunt US.
[One colleague darkly noted that there is another more capitalist-driven path to standardization - market consolidation. If one company simply buys out all the other methods, or if most of the competitors go out of business, all those differences will go away. The chaos of nonstandard values may accelerate the process, making it harder to have a profitable method or instrument. ]
Establishing reference methods and materials is the right scientific approach, but it is a costly and time-consuming process. We have examples of how to do this in the cholesterol and glycohemoglobin standardization programs, but other efforts in the US have lagged. Europe is leading the charge in this area and international guidelines, such as ISO 15189, are setting higher standards for quality (including traceability of calibration processes) than required in the US under CLIA. The US attitude of “quality compliance” is now starting to bite back!
Comments