If you were driving home and a cop ordered you to park somewhere, and then someone from the government asked for your saliva or blood while secretly testing you for alcohol use, would you feel like your rights were violated? Many Americans did when it happened to them, according to documents from a Freedom of Information Act request by Jalopnik and a closer look at the program's methods.

Thousands of Americans are still furious over the National Highway Traffic Safety Administration's $8 million, local police-aided roadblock alcohol and drug test surveys that was run in 60 cities last year.

Since the 1970s, NHTSA has run roadblocks in various states where drivers undergo breath tests and offer blood and saliva samples for the purpose of studying just how many people operate cars under the influence of drugs and alcohol. The program has been run for decades and has largely escaped controversy — until it came to Texas last year.

Documents from a Freedom of Information Act request filed by Jalopnik seeking complaints about the program following last year's outcry over roadblocks in the Dallas area, as well as media reports from the time, reveals an agency inundated with criticism over the program, and one caught seemingly unprepared to deal with the backlash.


Furthermore, while NHTSA has claimed the surveys were entirely voluntary, a closer look at the survey's methodology reveals stopped drivers were secretly tested anyway before they could give or deny consent.

Officials from NHTSA have not returned an email seeking comment.

NHTSA runs the roadblock studies, called the National Roadside Survey of Alcohol and Drug Use by Drivers, like this: local police stop motorists, order them into the roadblock area without talking to them, and then federal contractors ask them to participate in a survey about their drinking and drug use habits. The contractors later ask for samples of their breath, saliva or blood. Drivers pulled over were offered between $10 to $50 for their various bodily fluids.


In addition, NHTSA has admitted in their 2007 roadside survey results that while the surveys and tests were voluntary, survey-takers were secretly testing stopped drivers using "passive alcohol sensors" to help get as much data as possible before more detailed breath or blood tests could be done:

As part of the program, to protect survey participants and the public, it was important to know the extent of the drivers' drinking. To this end, a passive alcohol sensor (PAS), attached to the PDA with VelcroTM, was used to collect mixed expired air from approximately 6 inches in front of the driver's face [...] The PAS was held within 6 inches of the participant's face, and when the subject spoke, the interviewer activated the small electrical pump, which pulled in the exhaled breath from the participant.

The program has been implemented periodically for decades. NHTSA conducts these studies in order to determine the prevalence of drunk and drug-addled driving, and to develop new policies related to drunk driving education and enforcement. Groups like Mothers Against Drunk Driving and the Insurance Institute For Highway Safety have expressed support for the program.


While the number of drunk driving-related fatalities have plummeted over the last few decades, it remains a significant safety problem for motorists and pedestrians; in 2012, according to NHTSA's own stats, 10,322 Americans died in crashes involving drunk drivers, up slightly from the previous year.

If drivers at the roadblocks were found to be under the influence, they were not charged with crimes, unlike a sobriety checkpoint; they were instead "subjected to a safety protocol designed to dissuade his/her continued driving on that trip," according to the survey. Drivers are also given cab rides or even hotel rooms.

However, the invasive nature of the roadblocks drew considerable ire from civil liberties activists, police watchdog groups, and ordinary citizens alike. Though NHTSA claimed the stops were completely voluntary, numerous drivers felt differently, saying they felt "trapped" after police pulled them over and ordered them into parking spots.


(At least a few people also raised concerns that police and contractors were collecting samples to establish a DNA database of some sort, but there is no evidence that's the case.)

Here's one example, from Jalopnik's FOIA request for complaints about the program:


Unlike many states, sobriety checkpoints are not authorized by Texas's constitution, which is one probable reason the NHTSA program rankled so many, and even forced the police chief in Forth Worth to apologize for his department's role in testing motorists.

Here's how one civil rights advocate broke it down for CNN when the program came to Alabama last summer:

And Susan Watson, executive director of the Alabama chapter of the American Civil Liberties Union, called the use of deputies to conduct the survey an "abuse of power." Even though the survey is voluntary, people still feel they need to comply when asked by a police officer, she said. "How voluntary is it when you have a police officer in uniform flagging you down?" Watson asked. "Are you going to stop? Yes, you're going to stop."


DUI roadblocks — ones that can lead to actual arrests, not just surveys — have a contentious history in America. The Fourth Amendment protects Americans from unreasonable search and seizure, typically requiring probable cause that a crime has been committed before police can execute a search. But the Supreme Court in 1990 come down in favor of roadblock DUI stops, ruling that their usefulness to the state in reducing drunk driving outweighs the violation of privacy. Despite this, many states do not use roadblocks at all.

The first NHTSA roadblock survey was conducted in 1973 and has been performed every few years since; the last one was performed in 2007 and was done mostly at night but in the daytime as well. In 2007 about 10,000 drivers entered the roadside test sites and were deemed eligible for the study. Of them, about 9,400 agreed to give breath samples, while others gave blood and saliva.

The 2007 results claim 12.4 percent of nighttime drivers were "alcohol positive" in some way, which is scary to think about, although only 1.8 percent were found to be over most states' legal BAC limit of .08.


The good news is that's way down from the 1973 survey where a whopping 6 percent of drivers sampled were .08 or higher. On average about 10 percent of nighttime drivers tested positive for some kind of illegal drug, mostly marijuana and cocaine.

Still, despite drawing from a larger pool of people, the survey results note that participation dropped for the survey six years when compared to 1996; the "compliance rate" was 83.7 percent in 2007 versus 96 percent in 1996. They even say why this might have happened in the 2007 study's executive summary:

We suspect that the lower rates reflect national changes in the culture and attitudes toward survey participation (e.g., litigation concerns, nonparticipation rights). It is also possible that with the increase in computer- assisted telephone surveys and computer-generated telephone marketing calls, the public may have become more resistant to survey type activities.


But the biggest hurdle NHTSA faced? Getting local police to participate in the surveys at all, according to that year's summary:

The major barrier to carrying out this staged sampling system was obtaining law enforcement support for the survey. In some localities, city attorneys or law enforcement leadership with concerns such as potential liability and scare resources declined to participate. In these cases, substitution PSUs were obtained.

"PSU" means "primary sampling unit," the city, county or geographic area where the study took place. In other words, if NHTSA couldn't get local cops to play ball in one area of the country, they did it somewhere else.


And why wouldn't local police departments be inclined to stay away from participating in the program, especially in states like Texas where DUI checkpoints do not happen? After the blowback from the stops in Fort Worth, that city's police chief took to Facebook to apologize for his officers' involvement. Here's how he responded in one email:

And our FOIA request reveals that police in Fort Worth and Dallas were unwittingly put in charge of dealing with complaints from the public:


The results aren't in for the 2013 survey yet, but considering that people are now aware of domestic spying initiatives, and are generally more wary of government intrusions into their privacy, it's safe to assume participation could be even lower. CNN's report touched on the "timing" of the surveys with a critic on the other side of the political aisle than the ACLU:

Cliff Sims, publisher of the Alabama conservative blog Yellowhammer Politics, said the complaints are mostly because of the bad timing Bentley mentioned. But, he added, "I think it's also that it has a lot to do with a larger distrust of government and people feeling more and more like their privacy has been invaded.

"When you see that taken out of the online space, where it's not quite as tangible, and into the real, physical world, that's the kind of visible and tangible thing that people can latch onto," he said.


The Associated Press reported last week that at least one man who was stopped for the survey, Army veteran Ricardo Nieves of Reading, Pennsylvania, has filed a lawsuit against the Maryland-based contract company that administered the survey claiming violation of his Fourth Amendment rights.

Attorneys for the contractor, the Pacific Institute for Research and Evaluation, have asked a judge to throw out the case, claiming they made clear Nieves' participation was voluntary. Nieves refused the survey but told the AP his right to move about freely without government intrusion was violated when he was ordered into the roadblock.

Current concerns over privacy may be a large reason for the backlash in Texas, where numerous people emailed NHTSA with complaints which were uncovered by our FOIA request.


Texans weren't the only ones angered by the roadblocks. A state Senator in Tennessee recently filed a bill that would prohibit local cops' involvement in the NHTSA program. Here's what State Sen. Mike Bell told the Times Free Press:

"They may be seeking information that is valuable to know how safe our roads are. I have no issue with the company conducting a survey," Bell said. "The problem starts when you have law enforcement making the stop. You have a stop that is no longer voluntary. You have a stop that is a law enforcement action and it's being done without probable cause. That's wrong."

It's easy to see both sides of the debate over NHTSA's roadblocks. On one hand, nobody wants drunk or drug-affected drivers on the road; if this can help define new policies to keep them from driving, is it really so bad? Further, no drivers were ever charged with crimes. Instead, many were sent home or to hotel rooms, and it's possible that saved a few drunk drivers from the expensive process that comes with dealing with a DUI charge in court.


On the other hand, it's hard for NHTSA to justify calling the program "voluntary" when its contractors secretly did record drivers' breath alcohol levels before getting their consent. While the study has been done for years, it's perfectly understandable that many Americans will see this as one more step in the gradual erosion of their civil liberties. And the use of local police to herd drivers into the roadblocks would certainly make it seem less than voluntary to many drivers.

We eagerly await the full results of the 2013 survey to see who said "thanks, but no thanks" to this one.

Update: NHTSA responded to this story with the following statement:

"Each year, close to 10,000 people die in drunk driving crashes: 27 people a day, or one person every 53 minutes, according to data from the National Highway Traffic Safety Administration. To better understand the issue, the agency has regularly conducted its National Roadside Survey of Alcohol and Drugged Driving in communities across the country for over 40 years. The survey provides useful data about alcohol and drug use by drivers, and participation is completely voluntary and anonymous. More than 60 communities across the country participated in 2013, many of which participated in the previous survey in 2007. NHTSA always works closely with state and local safety officials and local law enforcement to conduct these surveys as we work to better inform our efforts to reduce drunk and drugged driving."


Photo credit AP