A recent scientific article from Forensic Science International confirms something that Seattle DUI lawyers already know, that breathing patterns can affect the result of a breath test in a big way. In "Influence from breathing pattern on alcohol and tracer gas expirograms—Implications for alcolock use", the Swedish authors conclude from their studies that the measurement of a breath alcohol result is greatly influenced by the timing of the test and the breathing pattern of the person taking the test. Specifically, shallow breathing or hyperventilation can lead to a lower breath test result. Conversely, deep breathing and holding the breath before blowing can lead to an increased breath test result.
There are a couple of points to be made here. First, the machine is supposed to be giving us an accurate measurement of what a person's actual breath alcohol content is. It should not be too high or too low based on how the person is breathing, especially since everyone breathes differently, and people who have been arrested are often crying, in shock, angry, etc, and not breathing in their normal breathing pattern. This is more proof that the number that the machine spits out is just a “guesstimation,” at best.
Second, police officers that know this information can use it to manipulate test results (just as a subject could). I am able to obtain and watch a video recording of the breath test in almost every Seattle DUI case, which I handle. It is very common to see police officers demonstrate to the subject to how to take a deep breath and hold it, before blowing in the machine. Whether the officer realizes that this can influence the test result, I don't know. But, it is very, very common. If nothing else, taking a deep breath and holding it tends to heat up the breath to be expired. The machine is calibrated to assume that the expired breath is 34 degrees Celsius +/- .2c. 34 degrees Celsius is 93 degrees Fahrenheit. The human body is, on average, at a temperature of 98.6 degrees Fahrenheit, or 37 degrees Celsius. Expired breath is expected to cool as it leaves the human body, thus the assumption of the lower temperature by the breath test machine. However, if the breath is warmer than 34 degrees Celsius, the test result will be skewed higher. Holding the breath will make it closer to the 37 degrees Celsius of the human body when it is expired.
The bottom line is that there are dozens of variables that can affect what the breath test result is at any given moment. These variables include breathing patterns, body temperature, partition ratio between the breath and blood, inherent margins of error in the machine, the inherent margin of error in the gas or solution used to calibrate the machine, burping, crying, hyperventilating, interfering chemical compounds, voltage fluctuations in the power supply to the machine, radio frequency interference from police radios, cell phones, etc. Every person's biology is different, but the machine treats everyone as if they were the same. Every person's breathing pattern is different, but the machine assumes that everyone is breathing the same. Alcohol in the blood is what makes people intoxicated, not alcohol in the breath. The machine has to make assumptions about the breath, and calculations and conversions based on those assumptions to estimate what the blood alcohol content is. The assumptions and estimates these outdated machines make leave too much reasonable doubt in my opinion as to what a persons actual blood alcohol level is and whether they are actually guilty of a Seattle DUI.