The general objective of this project was to determine where and how different types of DWI¹ enforcement systems fail in their mission to reduce alcohol-impaired driving and to suggest fixes for those failures. This project was concerned with improving the performance of governmental agencies that enforce laws limiting the blood alcohol concentration (BAC) of drivers of vehicles that operate on our nation's roads and highways. In this report we call the collection of these agencies in a given jurisdiction, and their procedures and resources, a "DWI enforcement system."

Specific objectives of the project were:

We envisage the Traffic Law System (TLS) as one of many societal systems that attempt to manage risk created by our Highway Transportation System (Jones and Joscelyn, 1976). Traffic crash risk is the particular domain of the TLS. In a positive sense, the TLS provides guidelines for the normal operations of the Highway Transportation System, and in a negative sense, it prohibits actions that create traffic crash risk and generates forces designed to control those actions. In general, its objective is maintaining crash risk at a level that is tolerable to society.

For many reasons, DWI enforcement systems often fail to maintain drinking-driving risk within tolerable limits. Such system failures are the result of a failure of the system to perform one or more of its constituent functions. Usually, the failures are not catastrophic, but merely result in reduced performance. For example, jurisdictions rarely cease to detect and apprehend DWI violators entirely, but only fail to detect and apprehend enough of them.

The reasons why these failures occur are of major concern in this project. The failure to detect and apprehend enough alcohol-impaired drivers may simply be due to too few police officers observing for cues to driving while intoxicated (DWI). However, failures are very seldom isolated but are interconnected with other failures. For example, the lack of officers observing for DWI may be due to a lack of command emphasis of DWI enforcement, which is due to a lack of public support for necessary resources, which is due to poor public information programs publicizing local alcohol-related traffic crashes and describing alcohol-crash risk compared with the risk of violent crimes. Thus, looking for chains of failures that define the failure modes of a DWI enforcement system is necessary. It is not enough just to look for individual failures.

The situation is made even more complex by the possibility of multiple failure modes behind a single system failure. Another failure mode resulting in a failure to detect and apprehend enough alcohol-impaired drivers may originate with poorly designed and time-consuming procedures for police-officer participation in adjudication proceedings. Such procedures could undermine police motivation to detect, apprehend, and process DWI suspects.

Finally, a system may be experiencing multiple system failures, with each system failure being the result of several simultaneously-occurring and mutually-reinforcing failure modes. For example, besides the failure modes contributing to the detect-and-apprehend failures described above, one might have several other failure modes resulting in a failure to impose effective sanctions to deter alcohol-impaired driving.

This project was concerned with identifying common failures and failure modes in DWI enforcement systems and in generating promising ways of dealing with them. Problems arising in the enforcement of laws dealing with driver impairment by drugs other than alcohol were not addressed explicitly in this project, although some of the same problems can occur in enforcing both alcohol-impairment and drug-impairment laws.

From the definition above, the DWI enforcement system is clearly concerned with a particular type of traffic-crash risk, that which is created by alcohol-impaired drivers. At the highest level, the formal functions of a DWI enforcement system are law generation, law enforcement, adjudication, and sanctioning, defined broadly as follows:

Besides the traditional functions listed above, a fifth, less formal, function is concerned with the dissemination of information among the components of the system and to potential DWI violators, among others.³

Many governmental agencies and institutions are involved in performing these functions. However, the DWI enforcement system has no "system manager" (because of the American doctrine of separation of powers), and has no "system specification" for describing what the system or any of its components should do. Actually, the DWI enforcement system is a "system of systems," each operating almost independently in some jurisdiction, but loosely bound by a common set of principles.

This project dealt with all of the functions of the DWI enforcement system as defined above, but was most interested in the enforcement function as affected by other functions of the system. The processes involved in the Law Generation function were not examined in this project, but pertinent BAC laws that must be "enforced" by the other three functions were of concern. In our analyses of these processes, we also looked for system failures that might be associated with laws specifying different BAC limits for different groups of drivers.

The project involved several tasks. First, we developed measures of the performance of DWI enforcement systems. Then, criteria were developed for selecting case-study sites at which the operation of various classes of systems could be observed. Criteria for selecting the members of an expert panel were also developed at this time.

We then developed a set of specifications of a traditional DWI enforcement system. The specifications were used as a framework for analyzing the systems and as a basis for comparing other types of DWI enforcement. Remaining tasks involved the identification of system failures and suggested fixes. All of this work was supported by ideas from the panel and by information gained in the site visits.

The expert panel was absolutely crucial to the conduct of this study, providing operational experience in all of the functional areas of the DWI enforcement system. The panel was not asked to reach a consensus on any particular issue, but merely to provide the individual members' opinions on those issues. It consisted of six members selected based on subject-area expertise and knowledge, willingness and ability to work and participate in cooperative group discussions, and representation of both national organizations and local practitioners. The panel helped in developing more detailed descriptions of common types of systems. The panel also helped in determining how and why these systems sometimes fail to operate as they were designed to operate, and in developing the fixes for the failures. Two, two-day meetings of the panel were held during the study. Other information about the operation of DWI enforcement systems was obtained through a series of telephone discussions with enforcement and adjudication staff in several states.

This report contains five chapters and one appendix. In Chapter 2, a detailed description of a baseline or nominal DWI enforcement system is presented as a basis for comparing other systems examined during the project. Chapter 3 describes DWI enforcement systems in three case-study jurisdictions. Problems being experienced in those systems, and some solutions being considered by various system actors are also discussed in Chapter 3.