Maybe the Instrument Is No Longer Linear Enough to Be An "8000C"
Updated: Oct 21, 2022
To explore the analytical variability of the 8000C as determined by the Alcohol Test Committee during evaluation of the 8000C prior to type approval.
To compare the analytical variability of the individual 8000C used in the subject tests for the client with the analytical variability of the type or class 8000C determined by the Alcohol Test Committee.
To suggest that the instrument in the matter before the Court may no longer be a member of that class or type because it is no longer capable of maintaining the analytical variability of the type or class.
To suggest that the instrument in the matter before the Court may no longer have the linearity of the class or type 8000C because it no longer exhibits the linearity established during evaluation of the type or class.
To establish that we have some evidence, that the linear kind of relationship, the created linear relationship that’s done by means of the calibration curve in terms of the – how the instrument works, in the case before the Court, we’ve got a possible interpretation that the relationship that normally should appear linear, the way that Ms. Martin found...is not linear.
To explain to the Court that creation of the calibration curve during factory calibration or re-calibration of an 8000C instrument takes something which is inherently non-linear and turns the instrument into a measuring device that can be used as if it were linear.
To suggest that if the instrument has lost that capability, over time, through drift, then the instrument ceases to be reliable and ceases to be a member of the type or class 8000C.
Q. Now when Ms. Martin did her evaluation and I think the actual results are reproduced in the Intoxilyzer training aid, if I remember correctly. A. In addition to her report to the A-T-C? Q. In addition to her... A. Evaluation, she also... Q. ...report to the.... A. ...published in the Canadian Society of Forensic Sciences Journal. Q. Yes. Why don’t we refer to that document, seeing as you’ve just mentioned it? A. Thank you. Q. You recognise this document? A. Yes. Q. You’ve seen it many times, I’m sure. A. A few times, yes.
Q. Page 23, the performance of two Intoxilyzer 8000C instruments was independently evaluated by two different laboratories, according to the A-T-C equipment evaluation procedures for approved instruments. Do you see that under “Method” page 23? A. Yes, first paragraph.
Q. Results of the testing, there’s a table at the bottom of page 26 under the heading, In vitro Testing Table 3. Accuracy and precision data for the evaluation are provided in Table 3. A. Yes.
Q. Now, just ask you to look at those results,and with respect to the instrument that was evaluated, the instrument was evaluated using solution at – and she’s been very exact in her target B.A.C. and entered it as 51.7milligrams per 100 mils is the target evaluation. Right?
A. That's correct.
Q. And again, the A-C-A and the A-B-A that we see referred to, that’s what we were talking about earlier, AC-A is the sequence air blank, calibration check, air blank;A-B-A is air blank, breath test, air blank.
A. That's correct. So, you’re using a simulator to determine the alcohol concentration.
Q. Right. Because in Canada, for these kinds of evaluations, we use a wet bath simulator.
A. Correct, as opposed to dry gas.
Q. Yes. So, with respect to testing at 51.7milligrams per 100 mils, she obtained a mean B.A.C., basically I gather that she took about 30 replicate analyses at each of the target concentrations, for each of A-C-A and A-B-A.
A. That’s what it says, yes.
Q. And so, the average, she would total up all of those values that ranged from 49 to 51 milligrams per 100mils, and then divide by 30 and she came up with the average of 49.9 using the A-C-A mode and 50.0 using the A-B-A mode.
Q. And the range of the results that she got,using the A-C-A mode, were 49 to 51.
Q. And the range that she got using the A-B-A mode were 48 to 52.
A. Yes, so as you can see, it’s slightly larger than it is when you have a controlled system that does it exactly the same each and every time.
Q. Right. And with respect to the 100 solution, which she was very exact in saying it’s 102 solution, using the same procedure, she received a mean B.A.C. of 100.2 using A-C-A and 100.2 using A-B-A. That’s an average.
A. Or mean, yes.
Q. Yes. And that was a range of values between 99 to 102 and – using A-C-A and a range of values using A-B-A of 97 to 102. Right?
A. Yes, correct. Again, the range is slightly larger, again, because of the difficulty of reproducing 30 breath tests when you’re blowing through the instrument,compared to when you’re putting it into a simulator and having the instrument do it repeatedly the same way every time.
Q. So, the results are consistent with the manufacturer’s specifications.
Q. And then she did an analysis of linearity of all of the results that she got, that’s at page 27.
A. Figure 1, yes.
Q. And she concluded that the relationship was linear.
Q. Now, she was fitting it to that same quadratic model that we talked about earlier.
Q. The instrument had – I suppose one way of looking at it, is that the instrument takes something which is inherently, in creating the calibration curve, the instrument takes something that is inherently non-linear and puts it into something that we can use as being linear. So, that the instrument produced linear results.
Q. So, the calibration curve is established at the factory, unless of course, I guess, there’s are calibration, but the instrument then has linearity at that point. The individual instrument has linearity at least in accordance with the experimentation by Ms. Martin as shown on page 27.
A. Correct. And again, this is done in a laboratory under controlled conditions.
Q. Well, in the best information that we’ve got, in the case before us, the same kind of linearity was not demonstrated during the periodic inspections of June 20th,2012, September 11th, 2013 and October 7th, 2014, for the instrument that we’re dealing with in the case before the Court.
A. That could be one interpretation, yes.
Q. But the point is, that we have some indication, we have some evidence, to use a legal term, that the linear kind of relationship, the created linear relationship that’s done by means of the calibration curve in terms of the – how the instrument works, in the case before the Court, we’ve got a possible interpretation that the relationship that normally should appear linear, the way that Ms. Martin found...
Q. ...is not linear.
A. That’s one interpretation, yes, that it’s reading low at the low end.
#linearity #calibration #drift
#analyticalvariability #typeapproval #linearity #linearizer #calibrationcurve #calibration