Over the past decade, states have increasingly turned to a powerful tool to stop drunken driving before it starts: miniature breathalyzers, wired into a car’s electronics, that prevent the engine from starting unless the person behind the wheel is sober enough to drive.
These devices, called ignition interlocks, have been remarkably effective. One study found that states mandating them for all drunken driving offenders had 15 percent fewer fatalities from alcohol-related car accidents.
But while interlocks have prevented thousands of crashes, they have also caused them.
In November 2017, Alexis Butler was backing out of a driveway in Arlington, Tex., when a speeding pickup slammed into her Toyota Camry. The driver, Blake Cowan, had been arrested twice the previous year for drunken driving, but he was sober at the time of the crash, according to police records.
To start his truck, Mr. Cowan had to prove his sobriety by blowing into an interlock device — a machine the size of a cellphone, wired to the car’s steering column. But to keep driving, he had to provide additional breath samples to show he hadn’t been drinking on the road.
Those checks, known in the industry as rolling retests, occur at random. They require the driver to lift a hand off the wheel, pick up the device and blow — hard — into its mouthpiece for several seconds. If the driver fails or doesn’t comply, the car goes into panic mode: its headlights flash and its horn honks until the driver turns off the engine.
Mr. Cowan had just passed a test, but dropped his interlock on the floor. Wanting it in easy reach for his next one, he distractedly fumbled for it, he told a police detective — and hit Ms. Butler’s car. She died a week later.
Nearly 350,000 people in the United States have interlocks, according to the latest estimate in an annual industry report, up from 133,000 a decade earlier. Thirty-four states — including, this month, New Jersey — require people with drunken-driving convictions to install the devices. That number will no doubt grow; other states are considering similar laws. Two United States senators are pushing legislation that would require all new cars to include a version of the technology by 2024.
But the devices’ potential to divert motorists’ attention has drawn little scrutiny, even as states clamp down on other forms of distracted driving, like texting behind the wheel or putting a phone to your ear.
A review by The New York Times of accident reports and lawsuits turned up dozens of examples of collisions in which the devices played a role. A Pennsylvania driver trying to complete a test blew so hard that he blacked out and crashed into a tree, nearly severing his left hand. Another in rural New Hampshire struck a telephone pole. And in California, a man attempting a rolling retest on a busy highway crossed the dividing line and hit another car, badly injuring a woman and killing her husband.
A Lucrative, Growing Industry
In 2005, New Mexico became the first state to require interlocks for everyone convicted of driving drunk. More than a dozen others soon followed, spurred on by a nationwide lobbying campaign by Mothers Against Drunk Driving.
The organization says its top legislative priority is getting all states to mandate interlocks for every drunken-driving offender. MADD has also called for automakers to include such technology as an option in all vehicles.
“The research showing that interlocks save lives is pretty overwhelming,” said J.T. Griffin, MADD’s chief government affairs officer.
That’s not the only reason they’re an easy sell: States don’t foot the bill. Drivers pay about $75 to have the devices installed in their cars, as well as a monthly monitoring fee that generally runs between $60 and $100. The annual cost of having a device is typically $1,000 or more.
That has created a lucrative industry. Smart Start, based in Texas, estimated its interlock revenue last year at around $150 million — nearly double its revenue from four years ago — according to its chief executive. Dräger, a German company that is among the leading manufacturers of breathalyzer machines, now makes twice as much money from interlocks in the United States as it does from its traditional breath-test business. Private equity firms looking to cash in have bought several interlock makers, including Smart Start, and are circling others.
When regulatory warnings about rolling tests have come up, Interlock companies have pushed back.
In 2006, the National Highway Traffic Safety Administration, the federal regulator in charge of setting vehicle safety equipment standards, began revising its 14-year-old guidelines for how interlock devices should work. A 2010 draft of the document said the agency “does not intend” that users perform rolling retests and said they should be performed while stopped on the side of the road.
The interlock industry and others objected, arguing that rolling retests were safe and, in any case, that it was impractical to expect drivers to pull over.
“All interlock vendors advise the client/user to pull off the road in a ‘safe’ place to take the retest,” the president of LifeSafer, a leading vendor, wrote in a letter to the N.H.T.S.A. He wrote that the “practical reality” of interlocks is that “99 percent” of retests happen while the vehicle is moving.
Another company, National Interlock Systems, wrote, “While it is possible that there have been traffic accidents while the test is being taken, we are not aware of any.”
And state officials from Colorado said rolling retests would actually be safer in many situations, like tunnels and “congested environments with tight lanes and limited shoulders.”
The regulator backed down. In its final guidance, published in 2013, it wrote that it was “very concerned about distracted driving” but would not specify how retests should be conducted. The agency also said that was “more appropriately a function for States and local jurisdictions.”
In a statement to The Times, the agency said the devices give “ample time” to pull over for a test.
The federal agency’s statement said it had not studied the issue of distracted driving and interlocks. States that mandate them have not, either.
David Kelly, the executive director of the Coalition of Ignition Interlock Manufacturers, said his group’s members were aware of very few accidents involving their devices — but he acknowledged that no one is actively looking for them.
“There hasn’t really been much research in this area at all, and part of that is because, from what we understand and the evidence we’ve seen, the distraction issue isn’t really that big of an issue,” Mr. Kelly said.
The Times worked with officials in North Carolina — one of the few states that makes most traffic collision records public — to see if interlocks were cited in police crash reports. The review went back 10 years, spanning three million crashes.
In 2017 — when about 11,500 of the state’s drivers had interlocks, according to an industry-funded study — there were eight crashes linked to the devices. In all, interlocks were involved in 58 crashes over the 10-year period. That is almost certainly an undercount: Not every crash report is logged in the state’s database, and many reports lacked searchable descriptions of the crash’s circumstances.
One driver told local police that he reached for his beeping interlock, missed a curve in the road and “woke up to someone saying he had been in an accident.” Two drivers rear-ended stopped cars during rolling retests. A fourth driver hit a sheriff’s vehicle. A fifth veered off the road and into a field, where he hit a calf.
Mr. Kelly said the crashes turned up in The Times’s review reflected just a sliver of the collisions caused by other distractions like changing a radio station, sipping a drink or toying with a cellphone.
“You’re never, ever going to eliminate all of the distraction risk inside a vehicle,” he said. “Every manufacturer tells customers they need to pull over for a retest and operate safely.”
People who favor the use of interlock devices say that any distracted-driving perils posed must be weighed against the far greater threat of drunken driving, while allowing offenders to get back on the road without driving on a suspended license.
“The interlock allows people to go back out and live their normal lives, while protecting everyone else on the road,” Mr. Leotta said. “It’s saving lives. I think the positive impact significantly outweighs any potential risks.”
In a letter to The Times, MADD’s president, Helen Witty, said interlocks had saved lives by preventing more than three million attempts to drive drunk in the past dozen years. She also suggested the accident reports supplied by officials in North Carolina were not necessarily accurate, because the drivers cannot be trusted to tell the truth.
“My two decades of experience with MADD has taught me to go beyond initial statements provided by a convicted drunk driver, particularly one who has once again gotten into some kind of trouble.”
Even when interlocks are used safely, sober drivers can end up in legal jeopardy.
The most advanced breath-testing devices are the machines found in many police stations. But they have had reliability problems, including software glitches, poor maintenance and human error. In recent years, courts have thrown out more than 50,000 such tests that were flawed.
Interlock devices use a cheaper and less trustworthy technology, the kind found in hand-held breath testers used in traffic stops. (Such tests are generally not considered reliable enough to be used in court.)
Mouthwash, toothpaste, breath mints and gum can trigger false positives. So can certain foods: Smart Start’s website warns drivers to avoid sugary snacks like doughnuts, cinnamon rolls and chocolate before blowing into their devices.
Chris Murray, a musician who was convicted of driving drunk in 2013, still doesn’t know why he failed a rolling retest.
Driving from Maryland to North Carolina for a gig in March 2014, he successfully completed several retests on his Guardian Interlock Systems device. Then, two hours into his trip, he failed one.
Mr. Murray was baffled. He’d had nothing but water and coffee to drink since he started his car. He pulled onto the highway shoulder, turned off his car and waited a few minutes to try again. He failed again. Half an hour later, a Virginia state trooper stopped to check on him — and Mr. Murray set out to prove he wasn’t drunk.
“I told him I want to do every sobriety test on the planet,” Mr. Murray said. The trooper later testified in court that Mr. Murray had showed no signs of impairment. Worried about getting to the gig on time, Mr. Murray called his bandmates for a ride and left his stranded car to be towed.
The failures triggered a report to his probation officer, who sought to revoke Mr. Murray’s probation and send him to prison. Terrified, he hired a lawyer and started barraging executives at Guardian with complaints about what he insisted were faulty tests.
The company didn’t budge. “Our Director of Training and Standards has reviewed the violations and agrees they are valid,” a Guardian representative responded in a letter reviewed by The Times. Company representatives did not respond to requests for comment.
Mr. Murray’s case went before a Maryland state judge. His lawyer called witnesses, including the Virginia state trooper, to back up his story. The judge ruled in Mr. Murray’s favor and ended his probation. He was left with about $4,000 in legal fees and other costs.
Ms. Butler never regained consciousness after Mr. Cowan crashed his truck into her car. Two months later, he was charged with manslaughter.
Mr. Cowan pleaded guilty and was sentenced this month to 16 years in prison. He is haunted by the crash and has twice tried to kill himself, according to his lawyer, Charles Smith.
Ms. Butler’s mother, Barbara Barr, said rolling retests should not be permitted. The devices, she said, should give drivers plenty of time to find a safe place to stop — and drivers should be required to pull over.
“We have laws against texting and driving,” Ms. Barr said. “They don’t have a law on interlock.”