Sometimes a criminal defense attorney can at times feel like Dorothy in the Wizard of Oz in that we are transported from the relative safety of home (the courtroom) to the weird world of Oz (the laboratory). There are unusual and often times conflicting phrases and words that seem to defy common sense. Sometimes, words and phrases are used interchangeably and with little apparent distinction. In this post we will examine a deconvolute some of these important terms.
Quality Control– We covered the concept of Quality Control (QC) before. QC is strictly speaking a process that is used to construct the calibration curve that our knowns are tested and then an unknowns are tested against. This is typically performed in the beginning of the run.
Standards/Controls– is a reference solution or test solution used for assessment of the performance of an analytical procedure. A rigorously tested and high quality known analyte at a certain concentration is known as a Certified Reference Material (CRM). CRMs are generally governed by ISO Guide 34:2009. NIST makes its own service marked brand of CRMs named Standard Reference Materials (SRMs). CRMs and SRMs should have statements of calibration (precision) and boas (accuracy).
Calibrator– Often it is in the beginning of a run. What makes a standard or control a calibrator is that it is placed before the unknowns are tested. The calibrators are run in a series over the hoped for linear dynamic range. The response from the testing of the calibrators is plotted on a signal versus concentration y-x axis graph. A line is drawn along the data points with the R2 calculated to determine if the response is linear. Calibrators are used to construct the calibration curve. It is the QC of a quantitative process. A calibrator is a solution hopefully from a traceable source and hopefully a CRM/SRM with a known amount (concentration) of analyte of interest that is hopefully pure and only contains that analyte of interest. It is placed within the batch of the run as part of the QC procedures to insure that the analytical instrument is detecting the known within an established stated and often times arbitrary range of values.
Verifier– Often it is in the middle of a run or at an end of the run. What makes a standard or control a verifier is that it is placed among the unknowns are tested after the run. It too contains a known analyte at a known concentration. It is placed within or at the end of the run to insure that the analytical instrument is detecting the known within an established tolerance throughout the testing of unknowns. Think of it as a check. If the testing method has a scheme where the laboratory places a verifier amongst the unknowns tested, then this is not a function of quality control, but rather an act of verification.
There is a big difference. QC data (where a series of calibrators are used over a range of concentraions) is used to establish a calibration curve. The data is inputted and R2 value is calculated and then the data adjusted to make it fit. Verification data simply tests at one point on the calibration curve already established, that data is then evaluated by a human, not a machine, and no adjustment is made to correct for bias if the verifier data result does not perfectly fit against the calibration curve expected result. The calibration curve is not altered based upon this new data point. Testing continues if the verification data is within an arbitrary range. Again, even if the verification data shows appreciable bias, nothing is done about it. Unfortunately, a lot of criminal laboratories in an effort to save money (which really is not that much) make their own in-house verifier solutions. I call this home brew or laboratory moon-shining. The difficulty with making it on your own is two fold: (1) You may make it wrong (impurities or imprecise or inaccurate concentration levels), and (2)it’s like the fox guarding the hen house in that you are trusting the laboratory to guard itself.
You must understand the difference between calibrators and verifiers.