We recently wrote about the Philadelphia Police Department's admission that four of its Intoxilyzer machines were improperly calibrated, which could result in the invalidation of hundreds of Pennsylvania DUI convictions from 2009 and 2010. Although court officials say that the inaccurate Intoxilyzer findings, which are generally only used to add more weight to other evidence in a drunk driving case, will generally not result in any overturned verdicts, the issue sheds light on Pennsylvania's lack of strict requirements for Intoxilyzer accuracy.

Intoxilyzers use infrared spectrometers to show the alcohol levels by examining how the molecules in the breath absorb infrared light. The device determines how much light is absorbed by the breath, compares that against known wavelengths, and converts it to the corresponding concentration of alcohol in the blood.

The devices must be re-calibrated annually to ensure that they are kept accurate. During this process, the official in charge of calibration will run a special liquid through the device in order to simulate blood alcohol concentrations of 0.05, 0.10 and 0.15 percent in order to check the accuracy.

Certification requires five tests of each simulated blood alcohol concentrations. If any device has an average deviation of more than 0.005 percent, it must be removed from service. In this situation, two units tested were off by 0.0054 and 0.0058 percent, which was technically over the allowable deviation limit. In addition, a third unit was off by 0.009 percent. However, a Pennsylvania judge still allowed all three of these units to be passed, as well as a fourth device which had significant errors in test procedure and paperwork.

Pennsylvania has the fewest and least strict regulations regarding Intoxilyzer accuracy of any state in the United States. It remains to be seen whether this will change in light of the recent issues.

Source: Philadelphia Inquirer, "Operator error in breath tests?" Don Sapatkin, 4 April 2011