Since 1973, five national surveys of U.S. drivers have estimated the prevalence of drinking and driving, and determined how this prevalence has changed over time. The National Highway Traffic Safety Administration (NHTSA) sponsored the first National Roadside Survey (NRS) in 1973 (Wolfe, 1974). The second NRS was sponsored by the Insurance Institute for Highway Safety (IIHS) in 1986 (Lund & Wolfe, 1991). The third NRS was jointly sponsored by the IIHS and NHTSA in 1996 (Voas, Wells, Lestina, Williams, & Greene, 1998). These three studies used the same basic methodology, which included a brief verbal survey and a breath sample to measure driver breath alcohol concentration (BrAC). The fourth NRS, sponsored by NHTSA, was conducted in 2007, with additional funding from the National Institute on Alcohol Abuse and Alcoholism (NIAAA), the National Institute on Drug Abuse (NIDA), and the National Institute of Justice. As in the prior studies, this study included a verbal survey and breath sample, but also added additional self-administered surveys and the collection of oral fluid and blood to determine the presence of other drugs in the driving population (Lacey, Kelley-Baker, Furr-Holden, Voas, Moore, et al., 2009). Self-report elements of the 2007 NRS were funded by NIDA and NIAAA. This fifth NRS was funded by NHTSA, with additional funding from NIDA and IIHS. This 2013-2014 NRS replicated the basic methodology used in the 2007 NRS, with protocol updates to include recent technological advancements and incorporate lessons learned during the 2007 study. Self-report elements of the 2013-2014 NRS were funded by NIDA and IIHS. A prescription drug survey, funded by NIDA, was added. All five studies were based on a national probability sample from the 48 contiguous states. The objective of this study was to estimate the prevalence of alcohol and/or other drugs in drivers across the country. Researchers interviewed more than 11,000 drivers to determine their alcohol concentrations and identify the presence of various over-the-counter, prescription, and illegal drugs in their systems. All interactions with subjects were voluntary and anonymous. Data included self-reported information, breath samples, oral fluid samples, and blood samples. The goal for each of the 60 sites was 125 oral fluid samples. Each site included 5 different locations, for a total of 300 locations. The sites were selected from the primary sampling units (PSUs) of the NHTSA National Analysis Sampling System/General Estimates System (NASS/GES). The NASS/GES PSUs are cities, large counties, or groups of counties from within four regions of the country and three levels of population density. Researchers recruited assistance from law enforcement agencies (LEAs) in these 60 sites. When law enforcement agencies declined to support the study, replacement sites were selected. Within each PSU, researchers randomly selected 30 specific square-mile grid areas and identified five data collection locations (a safe area to conduct the survey, with sufficient traffic flow for an adequate number of subjects). Drivers were randomly selected from the traffic flow. This multistage sampling system replicated the one used in the four prior NRSs (1973, 1986, 1996, and 2007). Researchers used a self-report screening instrument to detect alcohol use disorders (AUDs); and a similar instrument for drug use disorders (DUD), and the Drug Abuse Screening Test (DAST) examined potential drug abuse. Administration of these surveys was funded by NIDA and IIHS. New to the 2013-2014 NRS was the inclusion of a self-report prescription drug use questionnaire, funded by NIDA. These data will be reported through the funding partners. The protocol is summarized below. One or two law enforcement officers were present at each site for the safety of the drivers and the research teams. The law enforcement officers were not involved with interviewing drivers or any component of data collection. Large reflective orange roads signs indicated that the survey was voluntary. Vehicles guided into survey area: Randomly-selected drivers were guided into the research location, usually an empty parking lot. In some locations, the police officers assisted with traffic direction. Vehicles guided into individual research bay: A traffic director guided the vehicle into a specific research bay. Typically, six bays were set up; each was marked by orange traffic cones. Observational driver data: The data collector noted easily observable information about the driver(e.g., estimated age and race/ethnicity), and he or she recorded those data into an electronic tablet. Consent for interview: The data collector briefly explained the purpose of the study and that it was voluntary and anonymous. The data collector asked the driver for verbal consent for continuing the discussion. Researchers offered drivers financial incentives for completing additional parts of the survey. If the driver declined to participate, the data collector asked the driver if they were willing to provide an anonymous breath sample before the driver left the location. Drivers who were not willing, drove on. PAS reading: The data collector obtained an initial passive alcohol sensor (PAS) reading for the driver and recorded the result into the tablet. Survey interview questions: If the driver consented, the data collector asked a few questions regarding the subject’s general drinking behavior, driving patterns, and driving on that particular night (or day); the data collector entered this information into the tablet. Breath test: The data collector requested a breath sample from the driver. For drivers who consented, the sample was collected using a preliminary breath test (PBT) device, which masked the result, so neither the data collector nor driver knew the alcohol concentration. No identifying information was collected about the driver. Oral fluid test: The data collector requested an oral fluid sample from the driver. If the driver consented, the driver placed an oral fluid collection swab in his/her mouth for three to five minutes to collect approximately 1 milliliter (mL) of saliva. Self-administered questionnaire: While the swab was in the driver’s mouth, s/he completed selfadministered anonymous alcohol and drug surveys on the tablet while the oral fluid swab was in his/her mouth. a. Drug-use survey: Use of illicit drugs and, if the driver had used a drug, how long ago he or she had done so. b. Prescription drug survey: Use of medications and/or prescribed drugs/medicines. c. DAST survey: use of selected drugs, excluding alcohol and tobacco, during the past 12 months. d. DUD survey: Use of marijuana, cocaine, and pain killers. e. AUD survey: Use of alcohol, and to detect alcohol problems experienced in the past year. Passenger survey: If there was a front-seat passenger, researchers asked the passenger to complete a paper-and-pencil self-report survey while the driver was responding to the self-administered questionnaire. Payment: The subject received payment for completing the initial phases of the survey ($10 for oral fluid sample). Front-seat passengers who completed the passenger survey received $5. Blood sample: The data collector requested a blood sample. If the driver consented, the data collector led the subject to a nearby van, where a certified phlebotomist drew blood according to Occupational Safety and Health Administration (OSHA) standards. The subject received a $50 money order for providing the blood sample. Observational Vehicle Information: The data collector noted easily observable information about the vehicle and recorded those data (e.g., type of vehicle, number of passengers, and seat belt usage) into an electronic tablet. No personally identifiable information, such as license plate, driver’s license, or vehicle registration, was collected or recorded. Completion: The traffic director guided the driver from the research bay and back onto the roadway. Driver information card: The data collector completed this form to facilitate tracking and merging of data. Impaired driver protocol (IDP): If the data collector suspected the driver may have been drinking to any degree, or was otherwise impaired, a supervisor intervened and obtained a breath alcohol reading using an unmasked PBT device. If the driver’s BrAC was at or above .05,3 the research team ensured he or she got home safely. Slight modifications to protocol during data collection: In the beginning of the study, a small sample of drivers who initially declined to participate was offered an additional $100 incentive to reconsider participation. Halfway through the study, however, researchers stopped attempting to convert such drivers. Further, approximately two-thirds of the way through the study, researchers made a slight change to the protocol for using the passive sensors to accommodate feedback from law enforcement and the general public. An initial passive reading (collected prior to consent) was eliminated.4 Also, two-thirds of the way into the study, subjects were guided to the data collection location only by members of the research team. Additional variable message signage was also added — the sign indicated that the survey was paid and voluntary. Results This report presents only the methodology for the 2013-2014 study; results are in two separate reports - one on the prevalence of alcohol among drivers; and one on the prevalence of drugs among drivers. As indicated in Table 1, we selected more than 14,167 vehicles to participate in the 2013-2014 NRS; of these, 11,322 entered the data collection location, and 11,100 drivers were eligible to participate (e.g., commercial vehicles such as pizza delivery cars, emergency vehicles such as ambulances, drivers under the age of 16, and drivers who could not communicate in either English or Spanish were ineligible to participate). Almost 80% of eligible drivers participated in the survey, and because some drivers who declined to participate in the survey agreed to provide a breath sample, BrACs from the PBTs were available for 85% of the eligible drivers. Among eligible drivers, 71% provided an oral fluid sample, 67% completed a drug questionnaire and/or the AUD questionnaire, and 42% of drivers provided a blood sample. (Author/publisher)
Samenvatting