Executive Summary

Researchers compared the 2013 to 2017 fatal crash data from the KSP database to the crash data from the FARS database to check the consistency and accuracy of alcohol and drug involvement reporting for drivers in the KSP database. The indications of alcohol and drug involvement in the FARS database were used as the ground truth due to the inclusion of laboratory test results confirming the presence of alcohol and/or drugs in a driver’s system. For the five-year study period, the indications for alcohol involvement were 85.8% consistent and the indications of drug involvement were 66.2% consistent between the two databases. Of the fatal crashes where FARS confirmed alcohol involvement for a driver, KSP crash reports only identified 64.2% as involving alcohol. Of the fatal crashes where FARS confirmed drug involvement for a driver, KSP crash reports only identified 11.8% as involving drugs.

When comparing officer identification of alcohol or drug involvement to the confirmed presence either alcohol or drugs in FARS, consistency was 66.6% on average. Of the crashes shown by FARS to involve a driver testing positive for drugs or alcohol, officers identified 39.5% of these crashes as alcohol or drug involved. This increase in percentage compared to the consistency and accuracy of identification of drug involvement suggests that officers are aware that a driver may be under the influence of a substance, but they may not be indicating the correct substance.

The inclusion of drug concentrations found in a driver’s system in the FARS database could help make the distinction between drug presence that might affect driving abilities and drug presence due to a simple prescription, which may not affect driving ability. More training for officers to identify the signs of drug involved driving would also be useful to increase the percentage of drug involved driving identified by police.

Analysis of the crash characteristics from the KSP crash database showed that 93% of all alcohol involved fatal crashes occurred at nighttime, on a weekend, or as a single vehicle crash. Furthermore, alcohol involved fatal crashes were nearly two times more likely to occur at nighttime and three times more likely to occur on a weekend at nighttime than non-alcohol involved fatal crashes. Alcohol involved fatal crashes were also shown to occur more frequently in residential and rural areas than non-alcohol involved fatal crashes.

Alcohol and drug involved fatal crash rates were shown to be highest in Eastern Kentucky counties, with drug involved crash rates being higher, on average, than alcohol involved crash rates.

Report Date


Report Number


Digital Object Identifier



© 2018 University of Kentucky, Kentucky Transportation Center

Information may not be used, reproduced, or republished without KTC’s written consent.

The contents of this report reflect the views of the authors, who are responsible for the facts and accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the University of Kentucky, the Kentucky Transportation Center, the Kentucky Transportation Cabinet, the United States Department of Transportation, or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. The inclusion of manufacturer names or trade names is for identification purposes and should not be considered an endorsement.