Police routinely use breath-test devices (sometimes called “breathalyzers”) to determine whether a motorist was driving under the influence of alcohol. And prosecutors often use the results of these tests to prove DUI charges at trial. But are breathalyzers accurate?
Yes and no.
What a Breathalyzer Measures
When a person drinks alcohol, it eventually enters the bloodstream. As blood passes through the lungs, alcohol vapor escapes the body through breath. This vapor is called “breath alcohol.” Breath alcohol contains microscopic alcohol molecules that breathalyzers are designed to detect and measure.
Sources of Breathalyzer Error
Police use several kinds of breath-test devices. All of them are susceptible to error. Error can be caused by lots of factors, including:
- the device’s inherent margin of error
- physiological differences among drivers (“partition ratio”)
- radio frequency interference
- improper calibration of the device
- leftover alcohol in the driver’s mouth, and
- something that is tainting the breath sample.
Margin of Error
All breathalyzers have an inherent margin of error. In other words, even under the best circumstances, breath-test results may not be perfect. For instance, the margin of error for some breathalyzers is .01 percentage points. So, if one of these machines measure your breath alcohol at .08%, the actual amount could be anywhere from .07% to .09%.
Breathalyzers don’t measure blood alcohol concentration (BAC) directly—they estimate BAC based on breath alcohol. To compute BAC, breathalyzers multiply the breath alcohol measurement by a “partition ratio.” Breathalyzers use a preset partition ratio of 2,100 for this calculation. (See “Henry’s Law,” below, for more information on partition ratios.)
The problem is that not everyone has the same partition ratio. There are a number of physiological factors that affect partition ratios, including the driver’s:
- body weight
- breathing patterns
- hematocrit levels (portion of your blood volume that’s comprised of red blood cells), and
- body temperature.
So, partition ratios aren’t consistent for all people or even for the same person over time. Breathalyzers compensate for this variability by using a preset ratio that is actually lower than that of most people. The result is that most people benefit from the preset because the breathalyzer understates their BAC level. On the other hand, if a driver has a ratio that’s lower than the 2,100 preset, the breathalyzer will overestimate the BAC.
To curb the continuous battles over partition ratios during trials, most states changed their “per se” DUI laws to prohibit driving not only with an excessive BAC, but also with an excessive breath alcohol concentration. (BACs are typically determined by the number of grams of alcohol per 100 milliliters of blood, while breath alcohol concentration is usually measured in grams of alcohol per 210 liters of breath.) In the states that made this change—which include California and Arizona—partition ratio is no longer relevant for proving or disproving a per se DUI.
Radio Frequency Interference
Radio frequency interference (RFI) is another factor that can lead to inaccurate breath-test results. For example, the electromagnetic waves from a police radio can cause a breathalyzer to malfunction. To combat this issue, some breathalyzers have RFI detectors. If the breathalyzer detects RFI, it’s supposed to cancel any test in progress.
To work properly, breath-test devices must be calibrated regularly. A technician will normally calibrate the device by running a solution with a known alcohol concentration through the device. The device is then tuned until the reading matches the solution’s concentration. The laws of some states specify how often breathalyzers must be calibrated. In California, for example, breath-test devices have to be calibrated every ten days or 150 uses, whichever comes first.
While breath-test devices are designed to measure only breath alcohol—alcohol vapor from the lungs—sometimes alcohol from the mouth or stomach can make its way into the sample chamber. For example, if a person burps or vomits shortly before taking the test, the breathalyzer might detect more alcohol than is actually in the person’s breath. Alcohol lingering in the mouth or stomach has not yet made it into the bloodstream—or contributed to the driver’s intoxication—so including it in the measurement leads to artificially high test results.
To guard against mouth-alcohol contamination, police are supposed to observe a person for 15 to 20 minutes (called a “deprivation period”) prior to administering a breath test. During the deprivation period, the officer makes sure that the person doesn’t burp, vomit, or put anything in the mouth. Because alcohol evaporates quickly, lingering mouth alcohol should be gone by the end of the deprivation period.
Some breath-test devices use something called a "slope detector" to differentiate between mouth and breath alcohol. Breathalyzers measure breath alcohol continuously during a test. If the person has mouth alcohol, the reading will typically spike for a short period of time and then drop, creating an unusually steep down-slope. After the slope, the reading will plateau. The plateau is the person’s breath alcohol. Breathalyzers with slope detectors are supposed to report only the plateau measurement.
Tainted Breath Samples
Defense attorneys occasionally raise the possibility that the defendant’s breath sample was tainted either with the mouth alcohol of a prior arrestee or atmospheric vapors or fumes.
To prevent contamination from someone who police previously tested (and hopefully for sanitation's sake), officers are taught to replace the mouthpiece between tests. And to guard against contaminants in the atmosphere, most breathalyzers shoot an "air blank" through the sample tube. The air blank is supposed to clear out any lingering vapors or fumes.
Talk to an Attorney
If you’ve been arrested for or charged with a DUI, get in touch with a DUI attorney right away. An attorney can explain the law and procedure in your jurisdiction, help you understand what you’re up against, and let you know if there are any defenses that might apply to the facts of your case.