Are DUI Breath Tests Accurate?

There are a couple of ways to challenge the accuracy of the Intoxliyzer 8000 results. In Arizona, it used to be that you could get your own sample of the breath from the police at the time of the breath tests. This means they would have to give you a second sample of the same breath that you could have independently tested. The police have stopped doing that because it, frankly, was a pain in the neck to be doing all of that. They went to this new system where to make sure there was not a problem with the accuracy; you had to have two tests 5 to 10 minutes apart, known as replicate tests.

There was a case challenging legal and scientific reliability of the police using replicate breath tests alone without giving the person a second sample of the breath.  As a part of that decision, what the Court said is, well we know that you (the defendant) no longer can get your second sample of that particular breath to test however you can get all the Intoxliyzer 8000’s maintenance records and use the maintenance records to attack the reliability of the test results. In short, what the state is required to do now is keep the maintenance records and keep them accurately so we can use the records to challenge the accuracy of the tests results.

Back in the late 90’s, there was a major case involving the City of Phoenix where Phoenix was not accurately keeping those maintenance records. Basically what was happening is the quality assurance specialists, the one who did the accuracy checks on the machine, were saying that they did a test and the result was outside the accuracy range that it was their mistake, not that the machine’s and so it was not a fair test of the machine’s accuracy.  In fact, the quality assurance specialists did their testing procedure incorrectly and so they weren’t keeping records when a test result came outside the accuracy range.

There were in the end something like 600 cases that were part of the legal case of which 30 or 40 were mine.  The courts said that because the  quality assurance specialists had not kept accurate maintenance records so we could challenge the machine’s accuracy, the State  was not able to use the breath tests at trial. The state had to dismiss the 600 cases. The lesson the police learned was that the quality assurance specialists had to keep accurate records.

After the 600 cases were dismissed the law was changed to give the Department of Public Safety authority to set the standards as to how the maintenance records have to be kept.  One may to make sure the quality assurance specialists, who are part of the police department, do not make mistakes again is to have to police control and lessen the standards.  According to the Department of Public Safety standards the devices have to be calibrated at least every 30 days by a quality assurance specialist, who also have to run what’s called function and accuracy tests to make sure the device does various functions accurately. For example, does the device accurately detect mouth alcohol? Does it detect when a radio is nearby and it interferes with the testing procedure? Those are the function and accuracy tests. Those have to be done every 90 days.

The state’s required to keep records of those tests as well. When the police take the device out of service because of any particular problem and the most common would be that it’s reading outside of the accuracy ranges, it has to be taken out of service. The state has to keep all those records and we review those to see if we have an argument that it wasn’t working properly and accurately.
In addition to that, although the use is not like it used to be, they keep what’s known as C.O.R.B.A. records, which are records of how the device works every time it’s used. Every time a person takes a test, every time there’s calibration tests, all these records are preserved in a database, and you can go over those records and see how the device is working over a period of time.

If, for example, there’s a whole series of device reading outside of accuracy range when a person is taking the test, that information would be recorded in C.O.R.B.A. records. That information could be a basis to attack the accuracy and reliability of the device.  For example if for C.O.R.B.A. records revealed a number of failures in the calibration tests, those records can be used to challenge the accuracy of the he test results.

The police are not required to keep the C.O.R.B.A. records. Some agencies will do that, some won’t. For example, the Phoenix Police Department, does because after it lost 600 cases for not keeping accurate records , they decided it’s better to keep every record they can think of. Tempe does not keep C.O.R.B.A. records.

In terms of arguing about the accuracy and reliability of the breath test results what I like to point out is the key in terms of accuracy are what were the calibration checks that were run right before and right after the person took the test. What I like to argue is the police never test the calibration solution in the cylinder separately from the device. When the quality assurance specialist put a new cylinder on the particular device they do not first test the accuracy in the solution by first testing the solution by hook the cylinder up to another and see what the other device measures the solution as.  The quality assurance specialist just put the cylinder and the solution in it on the device without ever testing to make sure it the alcohol level in the solution is what it is supposed to be. I like that argument and have had some good results with it.

I like to make the argument that in fact the solution that’s being used to measure the accuracy of the device has never been independently tested to see what it is. What are they measuring it against? They, in effect, use the solution to say the device is working properly and if they get an answer within the acceptable range, .009 to .110, they’ll use the result to say that the solution is correct. Well, that’s circular logic. You’ve got to have an independent way of verifying that the alcohol in solution is, in fact, what it’s supposed to be since they’ve never tested it. That’s another argument I have made about the accuracy.

In addition to that another argument is about the accuracy of the tests in general.  Keep in mind that the according to the Department of Public Safety standards, the two replicate test results have to agree within .02% of each other or for example, .088% and .068%. A person taking the test can blow anywhere within that range and it’s still an accurate test. It’s not a question if the device doesn’t get sufficient volume of air because it has to get sufficient volume of air in order to be a valid test, in order to get numbers within .025 of each other. If the device doesn’t get a sufficient volume of air it will read the error message, “Deficient Sample”.

The same person with the same internal alcohol level can blow a .088% and a .068%, and there’s nothing really about the way an operator would administer the test that would narrow that range. Either of those possibilities is can occur and be a valid measure of the person’s actual alcohol level. That is another way be one way to attack the accuracy the results.

The third way to attack the accuracy of the results is to point out that the manner in which the person takes the test. Do they, for example, blow very hard on it and then very softly? That can affect the test results. Also the way the device is set to mimic a person’s internal alcohol level can be used to challenge the accuracy of the results for the particular person.  This is what’s called the partition ratio.

These devices are set up to say that if a person blows a certain quantity of alcohol from their breath into the device, it will correlate to a blood alcohol level that’s a preset standard by the machine. There’s a variance there for a person that could affect the results.

These are several other ways to challenge the accuracy of the breath results however the biggest one, is body temperature. In order to be a valid test, the person has to have a normal body temperature of 98.6 degrees or something plus or minus one degree. If for some reason the person has a fever, in effect the heated breath makes the breath test results higher than they should be.

If I can establish that the person did have a body temperature at the time of the test that can affect the test result, and that’s another method of attacking the test results.

For more information on Procedural Accuracy of Tests, a free initial consultation is your next best step. Get the information and legal answers you’re seeking by calling [number type=”1″] today.


Get your questions answered - call me now for a free phone consultation (602) 467-3680