Hundreds of recent drunken-driving arrests in Minnesota may be tainted because of a flaw in the breath-test device that is replacing one tangled up in a court challenge over its reliability.
The Bureau of Criminal Apprehension (BCA) spent $1.7 million for 280 of the new DataMaster DMT-G devices, which are designed to take two readings of a driver's blood-alcohol content at once, using two methods. Last month, the bureau told law enforcement agencies it had shut off one of the testing methods, using fuel-cell technology, until inconsistencies in the cells' durability are corrected.
That problem with the DataMaster doesn't affect the accuracy of the other kind of testing, using infrared technology, said a BCA spokeswoman. She said that the DataMaster, which has been in use since last fall, is accurate.
At least three defense attorneys involved in legal battles over the old breath-test machine, the Intoxilyzer 5000EN, contend that issues with the DataMaster call into question any drunken-driving arrests made with the device and that they have evidence to support their claim. They also raised concerns that officers received inadequate training with the DataMaster, and that the BCA rushed the device into the field.
"To use science to make accurate determinations, the science needs to be right so you are protecting people's rights," said attorney Ryan Pacyga. "I think this is ripe for litigation."
2,500 tests with DataMaster
The DataMaster, like the Intoxilyzer, tests suspects once they are brought to the police station. When a driver is initially stopped, officers use a different type of machine to take a preliminary reading. Only the test at the station is admissible as a basis to charge a driver.
In 2011, the DataMaster was used for 700 breath tests and the Intoxilyzer for 14,800, according the BCA's annual report. So far this year, another 1,800 tests have been done with the DataMaster. The southern half of the state was the first to see the device. By summer's end it will be rolled out in the Twin Cities and the rest of Minnesota.
Dual-technology testing isn't commonly used by law enforcement in the United States, but the BCA wanted a higher standard of scientific analysis for drunken-driving tests, said John Fusco, whose company, National Patent Analytical Systems in Mansfield, Ohio, built the device. It was difficult to combine the infrared and fuel-cell technologies into a single device because each has unique characteristics, and "we weren't a big proponent of it," he said. The bureau said its original request didn't require the fuel-cell function.
Jill Oliveira, spokeswoman for the BCA, said the bureau encountered inconsistencies with the fuel cell's shelf life, or how long until it needed to be replaced. The DataMaster still provides infrared technology, a proven, reliable way to determine impaired driving, she said.
"While we have concerns about the fuel cell's shelf life, we have no concerns about the DataMaster's reliability," she said.
The bureau and the manufacturer continue to work toward resolving the shelf life issue, she said. If it can't be resolved, the devices would continue to be used with infrared only, she said. The DataMaster was the highest-performing, lowest-cost device among those reviewed by the BCA, she said.
Fusco said his company is close to finding a solution but claims shelf life isn't the problem. Fuel-cell testing requires high humidity to retain its sensitivity, he said, and the cold weather seen each year in Minnesota sucks humidity from the atmosphere.
"We take the blame for not doing a good job of recognizing this when we designed the device," he said. "It didn't show up when we tested it and when the bureau did their testing before putting the device out in the field. The bureau doesn't deserve a black eye for this, and I don't believe there should be any problem with previous arrests standing up for police."
Even with the fuel-cell difficulties, Oliveira and Fusco agree that the DataMaster is far superior to the Intoxilyzer, which employs only infrared testing. The DataMaster gives real-time results, prevents other chemicals from adding to the alcohol result and is compatible with technology that allows charges and other information to be filed electronically. That significantly reduces the time officers need to complete paperwork for drunken-driving arrests.
The DataMaster also includes access to the computer source code that runs the device, to ensure its accuracy.
The Intoxilyzer's source code has been at the heart of a legal challenge to 4,000 drunken-driving and implied-consent cases. A district court judge ruled in March 2011 that the code contained errors but that it didn't cause inaccuracies in test results. Defense attorneys appealed, and the two sides are awaiting a ruling on the case by the state Supreme Court.
The BCA, which has stood by the reliability of the Intoxilyzer, moved to the DataMaster as part of a scheduled replacement plan, and the legal challenges were not a factor, Oliveira said.
Any DataMaster test results obtained before the fuel cell function was stopped are accurate, she said. There is a built-in safeguard that rejects the final result if blood-alcohol levels from the two tests deviate beyond a set benchmark. If the two results are within the accepted range, the device chooses the lower reading as the driver's alcohol level.
"The bottom line is these devices help law enforcement officers get impaired drivers off the roads and help keep our roads safer for the rest of us," said Oliveira.
Attorneys weigh in
Pacyga, along with attorneys Jeff Sheridan and Chuck Ramsay, has clients who have been charged with drunken-driving offenses based on DataMaster results. Whether a software or mechanical error is behind the fuel-cell issue, Pacyga said, he has to assume there is a problem with the entire device.
Ramsay said the DataMaster is a great improvement from the Intoxilyzer but wonders how the bureau can be so confident that test results aren't inaccurate.
Sheridan said the BCA should have worked out any kinks in the device before subjecting hundreds of drivers to charges that may end up being invalid. He also chided the bureau for providing only six hours of officer training on the DataMaster, adding that officers aren't provided with a training manual afterward. Lt. Eric Roeske of the State Patrol said no more training was necessary because the device is similar to the Intoxilyzer and simple to operate.
Sheridan countered that he interviewed an officer who had received the training and that his lack of competency was glaringly apparent.
"Will I make hay with all of this?" he asked. "You bet I will."
David Chanen 612-673-4465