LAUSR

Abstract Objectives The current study investigated associations between minimum jail sentences included in state DUI laws and self-reported drunk-driving in the US. Scant existing research has been conducted on this topic. The current study is the first to use Behavioral Risk Factor Surveillance System (BRFSS) data to investigate associations between minimum jail sentences and frequency of self-reported drunk-driving among adults who report drunk-driving. Methods 2020 BRFSS data were used. Analyses were limited to participants aged 18 years and older who reported at least one episode of drunk-driving during the past 30-days. Participants missing data on age, gender, race, income, binge-drinking, or state of residence were excluded from analyses. The final sample size was 3,732. Websites from official state legislatures and/or data from state motor vehicle departments were used to gather current information on DUI minimum jail penalties. A dichotomous variable was created that categorized all 50 states and Washington D.C. as either (i) having a minimum jail sentence for first time DUI offense or (ii) having no minimum jail sentence for first time DUI offense. Multivariable negative binomial regression analyses were conducted using SAS® v9.4. The dependent variable was the number of self-reported drunk-driving episodes. Results Among the sample of adults who reported drunk-driving, residents of states with no minimum jail sentence for a first-time DUI offense had a 32% higher incidence rate of drunk-driving episodes [RR = 1.32,95% CI:0.98–1.77] compared to their counterparts residing in states with a minimum jail sentence for first time DUI offenses. Similar findings were observed in multiple sensitivity analyses conducted. Conclusion There was a marginally significant [p = 0.07] higher incidence rate of drunk-driving episodes among residents of states with no minimum jail sentence compared to those in states with a minimum jail sentence for the first time DUI. This was the first study to use BRFSS 2020 data to investigate this association and only among participants who self-reported drunk-driving. We therefore provide an important contribution to the literature on this association by using the most up to date data and among a valid sample of drunk drivers. Further studies that provide stronger causal inference are needed.