Drink Driving under Vietnam’s Zero-Tolerance Law: Evidence from Police Roadside Checkpoints in Three Major Cities, 2020-2025

BACKGROUND: Vietnam introduced a zero-tolerance alcohol limit for all drivers in 2020 and subsequently intensified enforcement. However, recent data on drink driving and associated characteristics are limited. OBJECTIVES: To describe drink driving among drivers tested at selected police roadside checkpoints in Hanoi, Ho Chi Minh City (HCMC), and Da Nang during 2020-2025 and to identify demographic, vehicular, temporal, and weather-related correlates of drink driving. METHODS: Repeated cross-sectional breath testing was embedded in routine police enforcement from December 2020 through April 2025. Police selected checkpoint sites and times for operational purposes and used calibrated breathalyzers to measure breath-alcohol concentration (BrAC) on all drivers passing through the checkpoints. The study-standardized data form also recorded sex, approximate age, vehicle type, date and time, weather, and location. Prevalences of drink driving were calculated among tested drivers. Multivariable logistic regression with city fixed effects estimated adjusted odds ratios (AORs) for any recorded BrAC >0.00 mg/L using drivers with zero BrAC as the comparison group. RESULTS: Among 102,422 tested drivers, the prevalence of drink driving in the latest available round was 5.8% in Hanoi, 9.6% in Da Nang, and 37.9% in HCMC. Men accounted for 97.5% of drink driving cases, adults aged 25-59 years for 90.1%, and motorcyclists for 96.3%. Across most city-rounds, a majority of positive tests exceeded 0.10 mg/L and about half exceeded 0.38 mg/L. Male sex (AOR 3.55; 95% CI 3.29-3.84), age 25-59 years (AOR 2.48; 95% CI 1.95-3.18), and motorcycle use (AOR 5.70; 95% CI 5.36-6.06) were associated with higher odds of drink driving. Weekend testing was associated with slightly higher odds, whereas afternoon or night testing and rainy weather were associated with lower odds relative to their reference categories. CONCLUSIONS: Drink driving was substantial among drivers tested at selected checkpoints, particularly among male, working-age motorcyclists. Because checkpoint locations and times were selected by police for routine enforcement rather than through probability sampling, the findings may not reflect citywide prevalence or the causal effect of the zero-tolerance law. They support consistent high-visibility enforcement, standardized monitoring of tests and positive results, safer late-night transport options for motorcyclists, and targeted social-norm interventions.