The objective of this project is to identify and recommend strategies for improving statelevel data on the nature and extent of drugged driving in the United States by addressing the most significant barriers that impede state efforts to collect and compile such data. According to National Highway Traffic Safety Administration (NHTSA, 2012b) Traffic Records Program Advisory Assessment, states need accurate and reliable traffic records data to understand traffic safety problems, and to select and evaluate countermeasures to address the problems and ultimately improve traffic safety. Inadequate state data on drugged drivers limits understanding the extent and nature of the drugged driving problem, communicating it to the public, and measuring how it changes over time and in response to efforts to reduce it. (Berning & Smither, 2014; NHTSA, 2012a, MMUCC Guideline Model Minimum Uniform Crash Criteria Fourth Edition DOT HS 811 631; National Transportation Safety Board, 2012a, Reaching Zero: Actions to Eliminate Substance-Impaired Driving Forum Summary.) While drugged driving is receiving increasing national attention, especially as a result of state legalization of marijuana for medical or recreational use, most state data on drugged driving in its current form is of limited use for measuring and tracking drugged driving incidents, evaluating the effects of changing laws regarding drug use and driving, or improving our knowledge about drug use and driving impairment. In addition to the limitations of existing data, the relationship between the presence and levels of drugs in a driver’s body and their impairment or crash risk is further complicated by the number of potentially impairing drugs, the complexity of drug metabolism compared to that of alcohol, the variability of effects on individuals, and individuals using more than one drug or combining alcohol and drug use. Unlike with alcohol, drug concentrations do not necessarily relate to impairment, and drugs may be detectable after impairment has subsided. Threshold concentrations for drugs, similar to the 0.08 blood alcohol concentration (BAC), have not been agreed upon, and may not be feasible. (Government Accountability Office, 2015) Drugged driving is commonly defined as driving under the influence of or impaired by drugs other than alcohol. Some states define drugged driving as drug-positive driving, though the drugs included vary by state. For the purpose of this effort, “drugged driving” refers generally to driving with any detectable amount of illegal or potentially impairing amounts of prescription or over-the counter medications in one’s system, which includes driving while impaired by any of these drugs. “Drug-impaired driving” describes driving while impaired by a drug or drugs other than or in addition to alcohol, as differentiated from drugged driving, which includes drivers whose impairment is unknown (e.g., roadside survey subjects). Drugged driving data of interest include data on the prevalence of drugged and drug-impaired driving, drug-impaired driving citation and adjudication data, and toxicology data for drivers arrested for driving under the influence of alcohol and/or other drugs and/or involved in serious injury and fatal crashes. The focus of this project is on improving the quality of data related to drugged and drugimpaired driving. Efforts to enforce drugged driving laws generally, and barriers to such efforts, are outside the scope of this project. Improving our understanding of drug use and driving impairment, while important to combating drug-impaired driving and in need of additional research, is also outside the scope of this project. The improved data that would result from implementation of the recommendations in this report, however, should help guide drugged driving enforcement efforts and improve our understanding of drug presence/levels and crash risk. Issues not specific to collection of data on the presence and levels of drugs in drivers, e.g., funding of law enforcement generally, are clearly relevant to improving data on drugged driving and are important to this project, but are not the focus of this project and thus are only briefly mentioned and not discussed at length. This project employed a synthesis of literature to identify barriers to state efforts to collect and compile data on drugged driving sufficient to quantify the nature and extent of the problem at the state level, and recommendations to address those barriers. The draft synthesis document was then distributed to a thirteen-member expert panel of invited toxicologists, prosecutors, law enforcement agency leaders, safety researchers, and other traffic safety professionals. As a group, the panel represented those who must detect and cite/arrest drug-impaired drivers, test the specimens collected during the arrest process, prosecute offenders, and use the resulting data to support decision making on laws, policies, and safety. In order to identify barriers and recommendations, we searched for studies and reports on: *the limitations of existing sources of state data on drugged driving among the general population, suspected-impaired drivers, and crash-involved drivers; * challenges and issues in drugged driving enforcement; * guidelines and recommendations for drugged driving research. This effort was not intended to catalogue barriers state-by-state, though examples and exceptions to barriers are included where appropriate. The expert panel convened on July 9, 2015 to discuss the barriers and to evaluate, refine, and prioritize the recommendations. The meeting and prioritization was facilitated using a modified Delphi technique with the following components: * Discussion of each section of the literature review including a read-through of the previously provided literature summary and recommendations. This resulted in refinement of recommendations in which some were edited, some new recommendations added, and some were dropped from further consideration. Discussion of the recommendations focused on two main factors: how easy or difficult would it be to implement the recommendation, and, if implemented, how large an improvement would be expected on data completeness, accuracy, and availability. * Prioritization preparation including discussion of the process, review of all recommendations, and posting of the recommendations on paper sheets placed around the meeting room. At this stage, each panel member was given a set of stickers equal to approximately two-thirds of the number of recommendations upon which they were to vote. They were told they could apply their votes however they saw fit, including putting all votes on a single item; however, the facilitation team highly recommended avoiding too concentrated a voting strategy. The purpose and importance of the process and outcome were stressed at this point by reminding the panel members that the session was not aimed at unanimity, but at an expression of consensus that each panel member could support in their own work and among their own colleagues. * First vote in which panel members (with exception of those who abstained from voting as noted in Appendix) placed their stickers on the pages beneath the recommendations they believed were the highest priority. There was no discussion during this stage as the experts walked around the room indicating their preferences. The votes were tallied and the total number of votes was written on the sheet for each recommendation. *Post vote discussion in which the panel members reviewed each of the recommendations in order from highest to lowest vote tally. At this stage, the facilitator invited the panel members to make a case for (or against) any recommendation that they felt had received too few (or too many) votes, and to make proposals for combining recommendations which the panel could then either approve or deny. This process resulted in a final set of recommendations on which the panel could indicate their final priorities. * Electronic final vote in which panel members who participated in voting completed an additional voting process. Each panel member was allotted 13 votes to distribute among the recommendations and permitted to apply more than one vote to any given recommendation. Panel members provided their votes via email and the project team tallied the results, and rank-ordered the recommendations as reflected in this document. Recommendations with 10 or 11 votes were designated high priority, those with 6-9 votes as medium priority, and those with 2 or 3 votes as low priority. * Final review and approval in which each panel member was asked to “sign off” on the document and the indicated priorities of the recommendations. In the submittal email, the project team stressed that their approval was to indicate that they could support the documented priorities as reflective of the group’s discussion and conclusions. In this manner, this final report has been vetted by the group of experts assembled as a panel even while, individually, each may hold a different opinion on the precise order of the priorities. This document presents the revised synthesis of barriers and prioritized recommendations based on the discussion and voting by the expert panel, the members of which are listed in the Appendix. This document is organized into two main sections: Synthesis of Barriers and Expert Panel Recommendations. The barriers are organized into three broad categories of drugged driving data: (I) toxicology data from drivers who have been arrested and/or involved in crashes, (II) arrest/adjudication and crash data, and (III) data on the prevalence of drugged driving in the population generally (not limited to arrested and/or crash- involved drivers). The recommendations are presented in rank order and broken out into high, medium, and low priority based on the panel’s final vote. (Author/publisher)
Samenvatting