Program Profile: Checkpoint Tennessee

This is a statewide sobriety deterrence program that seeks to combat impaired driving and reduce alcohol-related car crashes and fatalities in the state of Tennessee. The program is rated Promising. The program was found to have a statistically significant reduction on the number of drunk driving fatal crashes and nighttime single-vehicle injury crashes, compared with preintervention trends.

A Promising rating implies that implementing the program may result in the intended outcome(s).

Program Goals/Target Sites

In the early 1990s, nearly half of all traffic fatalities in Tennessee each year were alcohol related (Lacey et al. 1999). Based on the success of Australia’s random breath testing (RBT) program (Homel 1990), Tennessee in March 1994 implemented a yearlong statewide sobriety checkpoint program called Checkpoint Tennessee. The goal of the program was to deter impaired driving and reduce alcohol-related crashes.

This impaired driving enforcement approach was a joint effort involving the Tennessee Department of Transportation, the Governor’s Highway Safety Office (GHSO), the Department of Safety, the Tennessee Highway Patrol (THP), and the U.S. Department of Transportation’s National Highway Traffic Safety Administration (NHTSA). This approach included checkpoints that were conducted on weekends in all 95 Tennessee counties; typically, a county conducted 10 to 15 checkpoints annually. The objectives of the program were to arrest impaired drivers at checkpoints, to get them off the streets, and to spread the message of zero tolerance for drunk driving statewide.

There were concerns over implementing sobriety checkpoints with regard to potential increased resources and cost (Lacey et al. 1999); therefore, another objective of Checkpoint Tennessee was to illustrate that the program could reduce drunk-driving fatalities while using existing personnel resources.

Program Components

The Checkpoint Tennessee program used a law enforcement approach in which checkpoints were scheduled across the state for a 12-month period. Checkpoints were implemented by THP in coordination with local law enforcement agencies. Three checkpoints were set up in at least four counties throughout the state every weekend. “Weekend blitzes” were conducted on five weekends out of the year, with checkpoints conducted in all 95 counties of Tennessee. These took place on the first weekend of the program in April 1994; on the weekends of Memorial Day, July 4th, and Labor Day; and at the end of the program, in March 1995.

At the checkpoints, all motorists were stopped to investigate for suspicion of impairment. If no evidence of impairment was found, they were given a pamphlet on driving under the influence and let go. If they were suspected of impairment, they were given further testing. Officers used several different types of equipment at checkpoints to detect driver impairment, including passive alcohol sensors in flashlights, video cameras, special lighting, cones, reflective vests, generators, signs, and floodlights. Passive alcohol sensors are flashlights with sensors to detect alcohol on the breath as the officer checks the eyes of the suspected drunken driver, and were an integral part of impairment testing at the checkpoints. All of these tools were used to gather evidence of impairment and determine probable cause to arrest impaired drivers. Standardized field sobriety tests were also used to detect impaired drivers.

Checkpoint Tennessee was heavily publicized in mass media outlets in the forms of TV, radio, and print public service announcements (PSAs), brochures, and billboards. Agreements were made with television stations to air the PSAs during primetime hours to maximize the impact of the message. Editorials were published in newspapers, discussing Checkpoint Tennessee and its goals. In addition, schedules for upcoming checkpoints were published in newspapers to make their presence known to the community. Billboards and radio ads were used to spread the message to motorists while they were driving; the billboards were placed on widely traveled highways in the state. Brochures outlining both the details of the program and the state’s drunk driving laws were distributed to the community, along with promotional materials, such as pens, pins, and cups.

Program Theory

A premise underlying the Checkpoint Tennessee program was that mass media outlets could have a strong influence on society (Lacey et al. 1999). Thus, advertising the program through media outlets was thought to increase awareness of the program and spread the message against impaired driving. Specifically, Checkpoint Tennessee operated on the idea that, if the program were advertised widely, people would be deterred from driving while impaired. Research finds that sobriety checkpoints can reduce the amount of alcohol-impaired driving and of deaths and injuries on the highway (Ross 1992).

Study

Lacey and colleagues (1999) used a one-group interrupted time-series analysis to determine the effects of the statewide Checkpoint Tennessee on the number of drunk driving fatal crashes and nighttime single-vehicle injury crashes during the 12-month intervention period (April 1, 1994, through March 31, 1995). For the study, Tennessee Highway Patrol (THP) officers conducted scheduled checkpoints on weekends in all 95 Tennessee counties. The intervention was intended to deter impaired driving and reduce alcohol-related crashes in a statewide initiative. Thus, there were no control counties in the State of Tennessee.

During the intervention period, 882 checkpoints were held. A total of 144,299 drivers passed through these checkpoints, and 773 were arrested for driving under the influence of alcohol (DUI) or driving while intoxicated (DWI). THP officers also arrested 201 drivers for drug violations, arrested 84 drivers for youth criminal violations, made 35 felony arrests, seized 49 weapons, cited 1,517 for safety belt or child restraint violations, issued 7,351 traffic citations, and recovered four stolen vehicles.

Data were obtained from submitted monthly reports by the THP and from the National Highway Traffic Safety Administration’s (NHTSA’s) Fatal Accident Reporting System (FARS; renamed as the Fatality Analysis Reporting System). Monthly reports were submitted every Monday from each of the eight THP districts, which contained reports of activity at each checkpoint from the previous weekend. Fatal crash data were retrieved from FARS, which included preintervention data (January 1988 through March 1994). There were approximately 108 monthly report observations included in the study analysis. To account for residual effects, data were examined after the intervention period until July 1996.

Outcomes of interest were the number of drunk driving fatal crashes and nighttime single-vehicle injury crashes. Drunk driving fatal crashes were defined as a fatal crash in which one of the drivers had a blood alcohol concentration (BAC) of 0.10 percent or higher determined by the driver’s BAC results (or the developed algorithm by NHTSA). A single-vehicle injury crash was defined as a less-severe crash, but one likely caused by an impaired driver.

The study authors did not conduct subgroup analyses. However, an autoregressive integrated moving average (ARIMA) time-series model was used to compare the number of drunk driving fatal crashes in Tennessee with the number of drunk driving fatal crashes in five nearby states. The results from this analysis are available in the Other Information section.

The total cost of the 2-year Checkpoint Tennessee project was $927,594. The U.S. Department of Transportation’s National Highway Traffic Safety Association provided federal funds of $452,255, and the State provided funds of $475,339. Federal funding covered public information, training and education materials, equipment, and program evaluation. State funding covered police salaries, publicity costs, and various other program expenses. The program did not use federal funding to hire new personnel or pay salaries of staff for the checkpoints, and instead used existing personnel resources. The reallocation of existing state resources demonstrated the feasibility of implementing this type of program in other police departments (Lacey et al. 1999).

The collaboration between the Tennessee Department of Transportation, the Governor’s Highway Safety Office (GHSO), the Tennessee Department of Safety, and the Tennessee Highway Patrol (THP) was imperative to the implementation of the Checkpoint Tennessee sobriety program. Specifically, the GHSO initiated the program through the development and planning stages, while the THP took initiative in the implementation of checkpoint activities. Central planning for the enforcement activities was coordinated by the Planning and Research Section of the THP. This resulted in a statewide plan and schedule for the implementation of the checkpoints for the full year, that was developed and distributed through the chain of command to a lieutenant in each of the THP’s eight districts responsible for the implementation of the schedule (Lacey et al. 1999).

Additionally, special training was necessary to successfully implement the checkpoints. Officers were specially trained to use the checkpoint equipment properly, especially the passive alcohol sensors. Procedures for checkpoints were done in accordance with Tennessee laws on checkpoint implementation, which included requirements for personnel and equipment.

To gauge public perception and awareness of the program (such as perceived risk of arrest for a DWI, support of checkpoints, and seatbelt use), researchers supplemented the study with opinion surveys. They distributed surveys at driver’s license renewal offices, conducted telephone surveys, and mailed comment cards. The surveys were administered at three different points (March 1994 [Wave 1], summer 1994 [Wave 2], and spring 1995 [Wave 3]) throughout the program’s implementation. Over the course of the intervention, the public supported the use of checkpoints (88.1 percent, Wave 1; 91.6 percent, Wave 2; 91.4 percent, Wave 3), were less likely to drive within 2-hours after drinking (17.4 percent, Wave 1; 15.6 percent, Wave 2; 14.1 percent, Wave 3), and had slight differences in their perceived risk of arrest for a DWI (47.0 percent at Wave 1; 43.5 percent at Wave 2; 47.0 percent at Wave 3).

Additional Outcomes

In regard to the autoregressive integrated moving average (ARIMA) time-series model, Lacey and colleagues (1999) compared the number of drunk driving fatal crashes in Tennessee with the number of drunk driving fatal crashes in five nearby states between 1988 and 1996: 1) Kentucky, 2) Georgia, 3) Alabama, 4) Mississippi, and 5) Louisiana. None of the five nearby states had a statistically significant reduction in the number of drunk driving fatal crashes that coincided with the Checkpoint Tennessee intervention. In other words, the decreased number of drunk driving fatal crashes in Tennessee were likely due to the Checkpoint Tennessee program.

These sources were used in the development of the program profile:

These sources were used in the development of the program profile:

Homel, Ross. 1990. “Random Breath Testing and Random Stopping Programs in Australia.” In Wilson, R.J., and Mann, R. (eds.), Drinking and Driving: Advances in Research and Prevention. New York, N.Y.: Guilford Press: 159–204.

Ross, Hugh Lawrence. 1992. “The Deterrent Capability of Sobriety Checkpoints: Summary of the American Literature.” Technical Report. Washington, D.C.: National Highway Traffic Safety Administration, DOT HS–807–862.