Novel Sobriety Program Cuts Mortality in Drunk Drivers by 55%

Batya Swift Yasgur, MA, LSW

March 07, 2023

A novel sobriety program that combines alcohol abstinence orders with frequent alcohol testing and rapid sanctions for noncompliance significantly reduces mortality risk in drunk drivers, new research suggests.

Results from a large study showed individuals arrested for driving under the influence (DUI) who participated in the 24/7 Sobriety program had a 55% lower probability of death within 5 years compared with DUI arrestees who didn't participate in the program.

"24/7 Sobriety is a program focused on reducing alcohol consumption among those involved with the justice system for alcohol-involved offenses," study co-author Beau Kilmer, PhD, McCauley Chair in Drug Policy Innovation, RAND Corporation, Boston, Massachusetts, told Medscape Medical News.

"It is unclear whether 24/7 will yield similar results outside of the Great Plains, but the peer-reviewed evidence is strong enough to justify piloting some version of this program in other parts of the US," said Kilmer, who is also the co-director of the RAND Drug Policy Research Center and professor of policy analysis, Pardee RAND.

The study was published online March 1 in JAMA Psychiatry.

A Leading Cause of Preventable Death

"Alcohol is the third-leading cause of preventable death in the US, with alcohol-impaired driving alone claiming 11,654 lives in 2020," the investigators note.

"To address this problem, some jurisdictions are using 24/7 sobriety programs, which combine alcohol abstinence orders with frequent alcohol testing (eg, twice-daily breathalyzers, monitoring bracelets) and swift, certain, and moderate sanctions for noncompliance (typically 1 to 2 nights in jail) to reduce alcohol consumption among those arrested or convicted for alcohol-involved offenses," they add.

"Our previous research found that 24/7 Sobriety reduced rearrest at the individual level and was associated with county-level decreases in arrests for domestic violence and repeat drunk driving," said Kilmer.

"Since the program reduces alcohol use and alcohol-involved crime, we wanted to see if the program influenced mortality — and, to our knowledge, this is the first peer-reviewed study using individual-level data to assess the effect of 24/7 on health outcomes," he said.

To investigate the question, the researchers used linked criminal history and mortality data for people arrested for DUI of alcohol in SD between 2004 and 2011, some of whom participated in the 24/7 sobriety program (n = 11,827 participants and 48,834 nonparticipants).

They examined the mortality data through 2016, allowing ≥ 5 years of follow-up for each individual.

The researchers used several statistical techniques in their analysis.

"With an RCT, people are randomly assigned to a treatment or a placebo, but in a quasi-experimental study such as ours, people were not randomly assigned to the 24/7 program vs no intervention," Kilmer explained.

For example, the judge might order an individual with a DUI arrest to participate in the program based on the severity of the alcohol problem — perhaps choosing individuals with more severe or less severe problems, depending on the perspective of the judge. "Thus, there is a potential for selection bias," Kilmer said.

The first statistical model used by the researchers was a naive Cox model, which did not account for selection bias. To address selection bias concerns, Kilmer and colleagues utilized additional techniques as part of an "instrumental variable approach."

They implemented two different approaches — a two-stage residual inclusion Cox model with a frailty parameter and a bivariate probit model — and the instrumental variable they used was "whether the program was available in the individual's county because in South Dakota this program was made available in different counties at different times," he said.

"These instrumental variable approaches allowed inference of the local average treatment effect among those predicted to participate in the program when it is assigned and enabled us to rule out concerns about selection effects," Kilmer reported.

Public Health Dimension

The naive Cox model yielded a conservative benefit of the estimated benefit of the 24/7 program, compared to nonparticipation in a program (hazard ratio 0.88; 95% CI, 0.77 – 1.00).

The two-stage Cox model, which addressed potential selection effects, produced a "substantially larger effect size," compared to the naive Cox model, and arrived at a hazard ratio of 0.55 (95% CI, 0.40 - 0.75). The survival curves that the model revealed demonstrated that 24/7 participants "consistently experienced lower risk of mortality than nonparticipants."

Results from the bivariate probit model found that those arrested for DUI who participated in 24/7 Sobriety had a 55% lower probability of death within 5 years compared to DUI arrestees who didn't participate.

"These findings add a public health dimension to the growing evidence that 24/7 sobriety improves public safety by reducing rearrest," the investigators write.

They caution that their approach "addresses individual-level observables and selection, but not other concerns, such as variation in county-level implementation," nor can they "assess the extent to which deterrence vs other mechanisms" drive the findings.

Kilmer noted that 24/7 Sobriety has been implemented in other states. "Our peer-reviewed studies of 24/7 Sobriety in North Dakota and Montana also suggest it is improving impaired driving outcomes," he said.

A Novel Approach

Commenting for Medscape Medical News, Keith Humphreys, PhD, professor of psychiatry and behavioral sciences, Department of Psychiatry, Mental Health, and Population Sciences, Stanford University, California, noted that "if an alcohol treatment system delivered this much benefit, everyone involved would be justifiably proud."

Humphreys, who was not involved in the current study, noted that the large health benefits of this justice-led program are remarkable.

He pointed out that this program, which is already implemented in about a dozen states, is also being implemented throughout England and Wales, "so it's clearly capable of going to scale."

Excessive alcohol use "continues be one of our country's largest drug problems, accounting for 1 in 8 deaths of people ages 20 to 64," Kilmer added. "If we hope to reduce excessive drinking and its negative consequences, we must try something different."

The research was supported by the National Institute on Alcohol Abuse and Alcoholism. Kilmer reports receiving grants from the National Institutes of Health during the conduct of the study. The other authors' disclosures are listed in the original article. Humphreys reports no relevant financial relationships.

JAMA Psychiatry. Published online March 1, 2023. Research Letter

Batya Swift Yasgur, MA, LSW is a freelance writer with a counseling practice in Teaneck, NJ. She is a regular contributor to numerous medical publications, including Medscape and WebMD, and is the author of several consumer-oriented health books as well as Behind the Burqa: Our Lives in Afghanistan and How We Escaped to Freedom (the memoir of two brave Afghan sisters who told her their story).

For more Medscape Psychiatry news, join us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.