The United States ‘Sober DUI’ Problem

In the U.S., drunk driving is a serious problem, with data suggesting at least 12,000 drunk driving fatalities in 2025, over 30 victims a day. Yet tests to discern a driver’s potential drunkenness are often inaccurate, leading to ‘sober DUIs’ and wrongful arrests. A 2023 medical journal study suggests that 49.2% of sober participants who were given placebos subsequently failed standardized field sobriety tests, highlighting the high potential for false DUI assessments.

In this study, we’ll consider the states that feature the most (and least) significant DUI arrest totals, and we’ll address the ‘sober DUI’ issue by looking at case studies and sobriety tests to uncover associated problems (including sobriety test limitations).

Let’s start with a look at some of the states in which false DUI cases have been recorded.

States Where Sober DUI Cases Have Been Reported

Sober DUI cases have been recorded in 22 states, including Arizona, Tennessee, and Hawaii. A sober DUI case is defined as involving the following.

• A driver with a blood alcohol concentration (BAC) level of 0.000
• No detection of alcohol or drugs
• Charges that were either dropped or never filed
• Wrongful arrest allegations.

Tennessee features the largest documented case volume. In 2024 alone, the state reported 419 ‘sober DUI’ arrests (2,547 such arrests have been made since 2017).

Hawaii represents the strongest recent case study. According to an Associated Press report regarding an ACLU Hawaii lawsuit, 127 drivers arrested by Honolulu police later recorded a BAC of 0.000 via breath or blood testing. Only 15 were given a traffic citation, while just three were charged with drug-impaired driving. Subsequently, Honolulu police launched a comprehensive review of drunk-driving arrests going back to 2021.

Arizona Case Studies

Arizona has also been subject to several high-profile lawsuits. In each case, the allegation involved a sober driver being wrongly arrested after an officer or Drug Recognition Expert (DRE) misinterpreted a medical condition as an indication of impairment.

One lawsuit alleges a California woman was wrongfully arrested by Phoenix police after being pulled over for allegedly running a red light.

The complaint referenced body-camera footage that showed the light was in fact green, all alcohol and drug tests were negative, and the fact that officers falsely determined the woman was THC-impaired based on her red eyes and dilated pupils. The lawsuit further alleged that DUI enforcement quotas were behind the arrest, an allegation Phoenix police denied.

Additional Arizona cases include:

• Surprise 

Jesse Thornton, a 64-year-old retired firefighter, was arrested when officers suggested he had bloodshot eyes and was unsteady on his feet. His breath test produced a BAC level of 0.000%, while his balance issues were later attributed to knee and hip issues. Jesse later filed a $500,000 wrongful arrest claim.

• Maricopa County

A 63-year-old home health aide was arrested after a DRE suggested they had poor balance. Blood tests later confirmed sobriety: work-related fatigue was identified as the likely reason for the misinterpretation.

Arizona Legal Issues

Arizona’s implied consent law (A.R.S. § 28-1321) means drivers must agree to chemical testing if an officer establishes probable cause. By refusing an initial test, drivers face an automatic 12-month suspension of their license. This means drivers often undergo a blood test to establish their innocence.

And yet, even when blood tests clear a driver and any criminal charges are dismissed, administrative license suspensions through the Motor Vehicle Division may remain on record.

This means innocent drivers may nonetheless have to spend months clearing their name and restoring their driving privileges.

Field Sobriety Tests

Field sobriety tests (FSTs) were initially developed in 1975 through research sponsored by the National Highway Traffic Safety Administration (NHTSA) and the Southern California Research Institute. The tests originally comprised six roadside tests.

• Walk-and-turn (WAT)
• One-leg stand (OLS)
• Horizontal gaze nystagmus (HGN)
• Finger-to-nose
• Finger counting
• Drawing on paper.

After trials, the NHTSA subsequently moved forward with three standard nationwide tests: horizontal gaze nystagmus (HGN), walk-and-turn (WAT) and one-leg stand (OLS). The results of these tests are admissible in court and are meant to help officers establish probable cause, as opposed to serving as definitive proof of impairment.

The Three Tests

Horizontal gaze nystagmus (HGN) measures involuntary eye movement while a driver follows a moving object. While carrying out this test, officers are looking for:

• Lack of smooth visual tracking
• Clear nystagmus at maximum deviation
• An indication of nystagmus before 45 degrees.

NHTSA’s original studies found that HGN had a 77% success rate when it came to identifying drivers with a BAC of 0.10% or above.

Walk-and-turn (WAT) involves drivers walking nine heel-to-toe steps, turning, and repeating the steps in the opposite direction. Officers are looking for balance issues, missteps, the use of arms for balance, pauses during the exercise, or starting the test prematurely. Two or more of these indications represent a 68% likelihood of a BAC of 0.10% or above.

One-leg stand (OLS) involves drivers balancing on one foot for 30 seconds while counting aloud. Officers are looking for swaying, hopping, putting their raised foot down, or using their arms to balance. Two or more of these indications represent a 65% likelihood of a BAC of 0.10% or above.

Validation Studies

NHTSA’s laboratory studies (in 1977, 1981, and 1983) reported the following accuracy levels.

• HGN: 77%
• Walk-and-turn: 68%
• One-leg stand: 65%
• HGN and walk-and-turn combined: 80%

Subsequent state-specific field studies revealed higher levels of overall correct arrest decision rates, including these examples.

• Colorado: 93%
• Florida: 95%
• San Diego: 91% (at a BAC of 0.08% and above).

Recent Research

A 2023 JAMA study measured field sobriety test performance among sober participants who received placebos. For those carrying out the walk-and-turn maneuver, the false-positive rate was 56%; for the one-leg stand, it was 37%; for both tests combined, it was 28%.

The study makes it clear that the FST classification of drivers is either ‘impaired’ or ‘unimpaired.” Because it was already understood that all placebo participants were sober, an ideal test would therefore produce a false-positive rate very close to zero. A false-positive rate of 56% in a binary test represents a performance rate no better than random.

What a DUI Test Can (and Can’t) Prove

Roadside observations, field sobriety tests, and toxicology tests all offer very different answers that are neither definitive nor interchangeable.

Roadside Observation

Documents the driving behavior an officer witnesses, with specific physical indications potentially establishing reasonable suspicion or probable cause.

Field sobriety performance

Measures physical and cognitive performance against standardized indicators, which, while potentially suggesting impairment, can’t establish it beyond doubt.

Toxicology test

Identifies the presence (or absence) of alcohol or drugs, but can’t definitively establish whether a driver was impaired while they were driving.

NHTSA data tells us that drug testing practices widely vary from state to state. The variance includes whether testing is performed, which substances are involved, and how samples are analyzed, all of which serve to limit the potential for state comparisons.

The data also suggests that limited testing panels and laboratories can produce both false negatives and false positives. Regarding cannabis, a positive result (including the detection of non-impairing carboxy-THC metabolites) can indicate prior use as opposed to current impairment.

Additionally, a negative result doesn’t definitively establish that a driver was unimpaired if the relevant substance was not tested for or was no longer detectable.

States Featuring the Most DUI Arrests

According to the NHTSA, a confirmed 11,904 people died in alcohol-impaired crashes in 2024: that’s 32 deaths a day, or one every 44 minutes, with higher rates estimated across 2025. Around 30% of all U.S. traffic fatalities involve drivers with a BAC level of 0.08 g/dL or higher.

While there are hundreds of thousands of DUI arrests made each year in the U.S., the Centers for Disease Control and Prevention (CDC) suggests they may represent just a fraction of impaired-driving incidents.

Using 2023 National Survey on Drug Use and Health (NSDUH) data, the CDC estimates that 15.5 million people (aged 16 and older) drove under the influence of alcohol during the previous year, 12.1 million drove under the influence of cannabis, and 2.2 million drove under the influence of other illicit drugs.

There’s a significant economic impact. The NHTSA suggests alcohol-involved crashes cost the United States around $68.9 billion every year, while a first DUI conviction can cost a culpable driver over $10,000 in fines, legal fees, and associated expenses.

In 2024, U.S. law enforcement agencies recorded 804,926 DUI arrests. Men represented 74.2%, with women representing the remainder, 25.8%. Here are the states featuring the highest and lowest DUI arrest rates during that year, with New York leading the rankings by a significant margin. (Note: a high DUI rate does not indicate that a state also features a high sober DUI rate.)

Top 10 DUI States

Bottom 10 DUI States

Arizona ranks almost exactly halfway in the rankings in 26th spot with a DUI arrest rate of 6.57.

Drowsy Driving and Work Fatigue Can Look Like DUI Impairment

If a driver’s tired, it can affect their driving performance in a way that closely resembles alcohol or drug impairment. And this can increase the chances of an exhausted, sober driver attracting the attention of a law enforcement officer.

According to 2024 NHTSA data, 644 people died in drowsy-driving crashes during the year. Such crashes usually happen between midnight and 6 am or during late afternoon, usually involve a single driver and a high-speed road, and often show no signs of braking.

The National Institute for Occupational Safety and Health (NIOSH) reports that extended wakefulness causes measurable impairment. Additionally, being awake for over 17 hours can produce an impairment effect equivalent to a 0.05% BAC; 20 hours of being awake can lead to an impairment effect of 0.08% BAC (the legal limit); and 24 hours of wakefulness produces effects comparable to 0.10% BAC.

Sleep deprivation also significantly increases the chances of a crash. AAA Foundation research found that drivers who sleep for between 6 and 7 hours a night are twice as likely to crash due to fatigue as those who sleep for eight hours or more, while drivers who sleep for fewer than five hours increase their crash risk by four to five times.

NIOSH also suggests that almost 5% of drivers in 12 states admitted to falling asleep while driving, with the Governors Highway Safety Association estimating that nearly one in five drivers has driven while drowsy.

Night-shift and extra-shift workers are especially vulnerable to fatigue-related driving, closely resembling DUI impairment. Even when chemical tests return negative results, drivers can still face lesser charges (such as reckless or careless driving) depending on the circumstances.

If you believe you’ve been wrongly suspected of driving while impaired, observe the following steps.

• Exercise your right to remain silent.
• Ask for a chemical breath or blood test.
• Keep hold of any evidence that may explain why you might seem impaired, such as your work schedule, time card, medical record, or any other document that confirms a sleep disorder or other medical condition.

Some states carry legal penalties for drowsy driving. New Jersey‘s Maggie‘s Law allows criminal charges if a fatigued driver causes a fatal crash, while Arkansas features legislation that addresses drowsy drivers who seriously injure or kill another person.

Misjudged Medical Conditions and Legal Medications

2023 CDC estimates suggest that 40.1 million U.S. people had diagnosed or undiagnosed diabetes, including 11 million undiagnosed adults. Low blood sugar can cause symptoms like blurry vision, shaking, fatigue, confusion, dizziness, seizures, and other behavioral changes; in all cases, this can look like impairment.

Some medical conditions and medications can cause DUI testing problems. For those with diabetes, elevated acetone levels can be misidentified as ethanol by some breath or urine tests.

Research in Jamaica involving type 1 diabetics found that 42.9% of tested individuals recorded breathalyzer readings above legal limits, despite many having consumed no alcohol.

Medication may also affect test results or driving ability. The U.S. Food and Drug Administration (FDA) suggests that many prescription and over-the-counter drugs can cause drowsiness, blurry vision, dizziness, slow reaction times, or compromised concentration. Some of these drugs include opioids, benzodiazepines, muscle relaxants, sleeping pills, antidepressants, and antihistamines.

Standard toxicology tests may wrongly suggest drug use when combinations of drugs are involved. There may be issues in the following cases.

• Ibuprofen (Advil, Motrin) can create THC false positives
• Amoxicillin has occasionally been linked to cocaine false positives
• Sertraline and trazodone may cause false indications when combined with amphetamines or benzodiazepines
• Pseudoephedrine and some diet pills may screen as amphetamines.

Additionally, substances that contain alcohol or similar drugs can affect breath testing results. Examples include cough syrups such as NyQuil, throat lozenges, oral gels, and asthma inhalers such as albuterol.

A 2013-2014 National Roadside survey found that 10% of daytime drivers tested positive for prescription or over-the-counter drugs. Clearly, some legal substances may complicate impairment assessments and toxicology interpretations.

The U.S. ‘Sober DUI’ Problem

We’ve established that field sobriety tests can show meaningful false-positive rates even among sober drivers, with cases in 22 states. One cited study found a 28% combined false-positive rate for walk-and-turn and one-leg stand tests, while officers erroneously adjudged almost half (49.2%) of sober placebo participants to be impaired.

NHTSA validation rates suggest an individual test accuracy of between 65% and 77%, indicating that these are probabilistic tools as opposed to definitive measures. Overall, field sobriety tests feature a serious risk of misclassification.

Numerous documented cases of 0.000 BAC arrests or charges involving sober drivers have been reported in states including California, Texas, Arizona, New York, and Florida.

Tennessee reports the highest documented volume (419 cases in 2024; 2,547 since 2017). Hawaii represents a key recent dataset: 127 Honolulu arrests between 2022 and 2024 later showed 0.000 BAC results, and only a small fraction resulted in a citation or DUI-drug charge.

If a driver’s tired, it can affect their driving performance in a way that closely resembles alcohol or drug impairment. And this can increase the chances of an exhausted, sober driver attracting the attention of a law enforcement officer

Arizona cases emphasize disputes over Drug Recognition Expert interpretations, medical conditions, and implied consent laws, which can initiate an automatic license suspension even before a case has been resolved.

Sleep deprivation of varying levels can suggest increasing levels of impairment and result in a false suspicion that a driver has consumed alcohol or drugs.

The same is true of conditions like diabetes, which can mimic impairment via hypoglycemia symptoms. And some medications (opioids, benzodiazepines, antidepressants, and decongestants) may impair driving or lead to false positives during drug screening, with a 10% positivity rate for prescription/OTC drugs found in a national survey.

Ultimately, standard sobriety tests can lead to wrongful suggestions of impairment, and further problems even once a driver’s inculpability has been established.

Without enhancements to drug-testing kits, more granular enforcement assessments to account for alternative explanations for impairment-indicative behaviors, and even a universal national standard featuring key driver safeguards for driver impairment assessment, such problems will continue to lead to multiple complaints and lawsuits.

A DUI arrest in Phoenix can put your license, job, and record at risk. What matters is what you do next. A DUI defense lawyer in Phoenix can step in to protect your rights and start building a strong defense from day one.

At Suzuki Law Offices, our Phoenix criminal defense lawyers bring decades of courtroom experience and a steady, confident approach to every DUI case. We’re always available: when you call, we answer. When you have questions, we explain. We treat our clients like family because your future deserves that level of care. Suzuki is your lawyer.