The absence of data on driving under the influence of alcohol in road traffic studies: a scoping review of non-randomized studies with vote counting based on the direction of effects of alcohol policies

Background
Data concerning driving under the influence of alcohol (DUIA) are frequently characterised by their lack of availability, accuracy, or reliability. This inherent limitation poses challenges in the examination of the influence of alcohol policies on road traffic outcomes. The study pursued a dual objective:

1) delineating the methodologies employed in evaluating road traffic outcomes attributable to alcohol policies in instances where DUIA data are absent, and

2) scrutinising the impact of alcohol policies under conditions of missing DUIA data.

Methods
A scoping review was conducted encompassing non-randomised investigations concerned with the assessment of road traffic outcomes associated with alcohol policies in scenarios characterised by the absence of DUIA data. Studies published in English or French between 2000 and 2021 were retrieved until November 2021. Databases such as MEDLINE, APA PsycInfo, CINAHL, and SocINDEX were scrutinised. Risk of bias within the included studies was evaluated utilizing the Quality Assessment Tool for Before-After (Pre-Post) Studies With No Control Group. The selection process, data extraction, and risk of bias assessment were carried out independently and in duplicate. Vote counting, predicated on the directional impact of alcohol policies, was employed as a synthesis method. The protocol for this review was registered in PROSPERO under the identifier CRD42021266744.

Results
A total of twenty-four eligible studies were incorporated into our analysis. With regard to the first objective, the majority of these studies employed uncontrolled interrupted time series designs as their methodological approach, with a primary focus on assessing road traffic fatalities resulting from night-time crashes. Typically, the rationale for the absence of DUIA data was not elucidated. Concerning the second objective, evidence emerged indicating an association between alcohol policies and a reduction in road traffic fatalities. Subgroup analyses failed to reveal any evidence of an association between methodological modifiers and the directionality of positive effects on road traffic fatalities.

Conclusion
Caution is needed in interpreting road traffic outcomes within the context of alcohol policies when DUIA data are unavailable. Heightened diligence should be exercised in enhancing the reporting of outcome assessments. Future investigations ought to address a multitude of methodological considerations, such as the utilisation of more granular data, well-defined interventions, and controlled study designs. To corroborate or recalibrate the associations identified in studies lacking DUIA data, it is imperative to compare our findings with those derived from reviews encompassing datasets inclusive of DUIA information.