Investigation of the Accuracy of Alcohol and Drug Involvement Reporting

Staats, William · 2019 · ROSA P / University of Kentucky Transportation Center

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study investigates the accuracy of alcohol and drug involvement reporting in Kentucky State Police (KSP) crash reports by comparing them against laboratory-confirmed data from the National Highway Traffic Safety Administration’s Fatality Analysis Reporting System (FARS). The research was motivated by significant inconsistencies between police assessments and toxicology results, which hinder effective countermeasure development and judicial sentencing for impaired driving. The primary goal was to quantify reporting errors, identify crash characteristics associated with impairment, and propose strategies to improve data collection. The methodology utilized fatal crash data from 2013 to 2017. Researchers linked KSP and FARS databases using spatial coordinates and temporal data, as the systems lack common identifiers. FARS served as the ground truth due to its inclusion of blood alcohol concentration and drug presence data. Crashes were categorized into six groups based on the presence or absence of alcohol, drugs, or either substance in FARS records. The study analyzed crash-level attributes (time, location, collision type) and driver-level attributes (age, gender, restraint use, human factors) from KSP records to determine correlations with impairment. The findings reveal substantial under-reporting of impairment by law enforcement. For alcohol involvement, KSP and FARS data were consistent in 85.8% of cases; however, police failed to identify alcohol in 35.8% of crashes where FARS confirmed its presence. Drug involvement reporting was significantly less accurate, with only 66.2% consistency. Police identified drug involvement in just 11.8% of crashes where FARS confirmed drugs were present. When combining alcohol or drug involvement, consistency was 66.6%, with officers identifying impairment in only 39.5% of confirmed cases. This suggests officers often recognize impairment but misidentify the specific substance. Analysis of crash characteristics showed that 93% of alcohol-involved fatal crashes occurred at night, on weekends, or involved single vehicles. These crashes were nearly twice as likely to occur at night and three times more likely to occur on weekend nights compared to non-impaired crashes. Geographically, Eastern Kentucky counties exhibited the highest rates of alcohol and drug-involved fatalities, with drug-involved rates averaging higher than alcohol-involved rates. The study concludes that inaccurate reporting impedes the ability to target enforcement and treatment programs effectively. The authors recommend enhanced training for officers to distinguish signs of drug impairment from alcohol impairment and suggest using crash attribute surrogates—such as nighttime, weekend, and single-vehicle occurrences—to flag potential impairment in cases where toxicology is unavailable. Additionally, the inclusion of drug concentration data in reporting systems could help differentiate between impairing substances and non-impairing prescriptions.

Key finding

Police officers identified only 11.8% of fatal crashes where laboratory tests confirmed drug involvement, compared to 64.2% for alcohol-involved crashes.

Methodology

dataset

Sample size: 3380

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (10 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence
• illicit drugs
• alcohol
• sex gender
• polydrug
• naturalistic crash near crash

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes, observational prevalence
• Methodological Resource: dataset resource