3 Consecutive Annual Inspections Reveal a Trend Towards Unreliability
To obtain an admission from the government scientist that a review of the documents from 3 annual inspections reveals a trend such that indications in the lower portion of the measuring interval are reading low and indications in the upper portion of the measuring interval are reading high, in other words, the response of the instrument to known levels of alcohol standard has drifted such that there has been a trend in change in the accuracy of the instrument.
To consider the alternative explanation that the inspectors are not following the same standard operating procedures during inspection from one annual inspection to the next.
To obtain an admission that further disclosure or production of information from the inspector on the annual inspections is necessary to properly assess these anomalies.
To obtain an admission that the inspection procedure and use of the instrument continued even though the results were outside the low end of the range / tolerance.
Q. And so, in terms of results, the measurement results, that we are likely to receive from that instrument are less reliable. A. They’re going to be lower, yes. Q. All right. So, let’s take a look on September the 11th, 2013, page 34. A. Assuming that’s what the problem is, it’s the instrument response, or was it a problem with the simulator and how the process was done of determining those standards. Q. All right. A. Again, that’s something you’d have to speak to Steve Marsh about.
Q. Right. Or, perhaps we can see if something similar occurs the next year when the instrument is inspected. Page 34. Same set of questions. The response at 50 milligrams per 100 mils we – they received a 44 and a 43.
A. Correct. Q. Are those within 3 percent of the target value of 50? A. No, they are not. Q. Are they within 5 percent of the target value of 50? A. No, they are not. Q. Are they within 10 percent of the target value of 50? A. No, they are not. Q. And with respect to the solution at 100 milligrams per 100 mils, with respect to that solution, those results, are they within 3 percent of – of 100 milligrams per 100 mils? A. No, they are not. Q. But they are within 5 percent of the target concentration of 100 milligrams per 100 mils? A. And the 10 percent, yes. Q. Yes. And with respect to the 300 solution, it’s now reading high; 308 and 307. Have I got that right? A. Correct, yes. Q. So the response now in the instrument has shifted such that some parts of the response of the instrument across the range are too low and some parts of the response of the instrument across the range are too high. A. That’s what it appears like, yes.
Q. All right. So, now October the 7th, 2014. Page 40. All right, now let’s take a look please at, again, the same set of questions, at 40 milligrams per 100 mils there are readings of 35 and 34. A. So, just to make a note, that it appears that the concentration of the low alcohol standard used has
changed from 50 to 40 milligrams of alcohol in 100 millilitres of blood from June 12th – sorry, June 20th, 2012 to October 7th, 2014. Is that correct, as I see that? Q. That’s – that’s as I read it as well. A. Okay. Q. All right, so now with respect to at 40 milligrams per 100 mils the numbers that are produced there of 35 and 34, are those within 3 percent manufacturer’s specifications for the instrument? A. No, they are not. They are also reading low. Still reading low. Q. And with respect to the 100 milligrams per 100 mils, are they within 3 percent? A. Yes, they are, and within – obviously, 5 and 10 percent. Q. And with respect to the results at 300 milligrams per 100 mils, again, they are reading slightly high. A. Yes. Q. But within – within 3 percent of the target value. A. Yes. Q. So, it appears that we have a series of at least – well, three different periodic inspections where we have – we seem to have a trend of the instrument reading low when tested at 40 or 50 solution, reading close to normal at 100 solution and reading slightly high at 300 solution. A. That’s what it appears like, yes. Or, the instrument hasn’t shifted or changed in response, but how the procedure was done, and the consistency with which it was done, may have changed from inspection to inspection. Again, that’s something you would have to ask Steve Marsh.
Q. I see. Thank you. A. But in all those cases, the instrument at the low end is reading low, either at 100 or at 50 or 40. Q. Yes. A. Or it’s reading accurately, but again, how the procedure was done could be the other explanation, ‘cause again, it’s very difficult to provide samples into the instrument using a simulator. Q. Yes. A. I’m not sure if you’ve ever done that, but it requires a lot of force to do that. Q. Yes. A. And so there can be higher variability associated with the numbers. Q. All right, just.... A. But that’s not an explanation for why the procedure continued even though the results were outside the low end of the range.