Checking Analytical Variability and Accuracy of the Instrument? Where?
Updated: Oct 10, 2022
To explore the lack of a lab setting when accuracy and precision of the instrument are checked on periodic or annual inspection.
To explore the lack of Standard Operating Procedures when accuracy and precision of the instrument are checked on periodic or annual inspection.
Q. I see. All right. A. Now, if you wanted to look at the analytical variability and accuracy of the instrument. Q. Yes. A. Then you would have to go look at these here on page 43 of Exhibit 12.
Q. All right, so you’re back to Exhibit Number 12. A. Right. Q. I’ve better make sure I’ve got my copy of that document. Yes. A. All right. So, what we have here, is a wet bath.... Q. You’re looking at page number... A. Forty-three. Q. ...43, let me just find that in Exhibit 12. Yes. A. Right. So, we have the ability of the instrument to accurately measure the blood alcohol concentration from – blood alcohol concentration – the alcohol standard concentration that it should get an expected result of 100... Q. Yes. A. ...milligrams of alcohol in 100 millilitres of blood. But as we can see, there’s considerably lower numbers than that. Q. Yes. A. And we know that in the field that this can happen compared to in the laboratory.
Q. Yes. A. All right? And we also know that systematically, the Intoxilyzer, over the eight years now that it’s been in use in this province, that the calibration checks that tend to occur with that instrument, tend to be lower than the actual result.
[Comment: How is that something good? Isn't something wrong with that statement?]
Q. That sounds like drift. A. No, it’s simply a matter of all the other factors that are out there, but also the pump with respect to the Intoxilyzer 8000C. It is not as strong as the one in the Intoxilyzer 5000C and so, therefore, the results that we see in these calibration checks tend to be lower than the expected value. It doesn’t mean that the instrument is drifting or that the calibration has – is less. It just simply means that these are the results that are obtained.
[Comment: Doesn't that mean that 8000C cal. checks are inherently unreliable?]
...Then you have the IT- P next to it, which shows that – so the I-T-P again, is where the instrument does an internal electronic check of the calibration of the instrument, as opposed to the wet bath simulator cal-checks which measure externally, right? So, it’s using a verified standard to externally check the calibration of the instrument. This is an internal electronic check of the instrument to see whether or not it can mimic 100 milligrams of alcohol in 100 millilitres of blood by changing the voltage going to the light source so that it can mimic the decrease in the intensity of the signal that is reaching the detector that would be the same as if someone had put a sample of breath into the instrument that is exactly 100 milligrams of alcohol in 100 millilitres of blood. And what we have here is, we have a result of 98.8 and then the standard deviation of those results.
[Comment: And when do the ITPs ever get re-calibrated? Don't they also suffer from uncertainty growth if there are not short re-calibration intervals? See FOI e-mail re re-calibration of ITPs]
And of course, the same applies to the calibration checks done using the wet bath simulator. The result is 96.5 and the standard deviation, which is the
variability or the precision is listed as 0.7071 and 0.7888. Q. Where did they conduct that? A. This would’ve been done at the detachment or it says, in the RIDE truck actually. Q. Says it was in the RIDE truck. A. Yeah. Q. So, [the local] Police, they only – the most recent piece of data we’ve got on this instrument, for 2014 – I mean, ‘cause we don’t yet have the data for – on the 6th of March, 2015, the 14th of April, 2015 or 20th of May, 2015. But the last data that we’ve got of one of these calculations of accuracy and precision using the stability test function of the instrument, the last one that we’ve got [the local] Police was doing that in the RIDE truck. A. That’s where the location is designated. I would assume that’s where it was either done or the location isn’t correct. But if there was any other wet bath calibration checks done or I-T-Ps done, that would’ve been in the downloaded data that was provided. There’s no physical hard copies in the papers I was provided but it would’ve been provided in the downloaded data.
Q. Let’s look at page 38 for 2013. There’s [the local] Police doing their calculations of accuracy and precision using this approach that your – and this is part of the policy of the Centre of Forensic Sciences, that this is how you check accuracy and precision? A. This would be one way of doing it, yes. Q. Right. That’s – that’s the – that would be the recommendation of the Centre of Forensic Sciences, that this is the way that you do it? A. There’s no recommendation from the Centre of Forensic Sciences that this is how it’s supposed to be
done. There’s no recommendation at all with respect to what parameters should be checked on the instrument during an inspection and how those should be done, but this is one way that the police service can do that.
[Quaere: Issue is lack of metrological supervision and control in Ontario?] Q. In a RIDE truck. A. That’s what it says. Again, I can’t confirm that location. I’ve testified on numerous cases where the location of the test was actually wrong in comparison to where it was actually performed, where an instrument was moved to a hospital, but the location was listed as a police detachment.
[Comment: How is this possibly good science? Don't we need metrological supervision and
metrological control in Canada?]
#analyticalvariability #inspection #annualmaintenance #precision #accuracy #SOPs