Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: J Safety Res. 2021 Oct 27;79:376–382. doi: 10.1016/j.jsr.2021.10.004

Alcohol-Related Deaths Among Young Passengers: An Analysis of National Alcohol-Related Fatal Crashes

Eduardo Romano a, James Fell b, Kaigang Li c,e, Bruce G Simons-Morton d, Federico E Vaca e
PMCID: PMC8640369  NIHMSID: NIHMS1750672  PMID: 34848017

Abstract

Introduction:

There is consensus that riding with an impaired driver (RWI) constitutes a major threat to public health. The aim of this study was to characterize the factors contributing to the motor-vehicle deaths of 15–20 year-old (y/o) passengers that RWI with a peer.

Method:

Secondary analyses of the 2010–2018 Fatality Analysis Reporting System. 5,673 passengers aged 15–20 y/o killed while riding in passenger cars with a driver aged 21 or older, 3,542 of these drivers also aged 15–20 y/o. Analyses were conducted between October 2019 and December 2020.

Results:

Sixty-three percent of the young passengers were killed while riding with a driver 15–20 y/o. Of these drivers, 26.8% had a blood alcohol concentration (BAC) >0.00 g/dL and 77.1% had a BAC ≥0.08 g/dL. Compared with those occurring during the day on weekdays, fatalities of young passengers who RWI with a peer driver with a BAC ≥0.08 g/dL often occurred on weekend nights (OR=8.2) and weekday nights (OR=5.2), and when the passenger and driver were both male (OR=1.8). Race/ethnicity was not a significant contributor to RWI fatalities.

Conclusions:

Most 15–20 y/o RWI fatalities occurred on weekends, at night, when the driver was a young peer with a high BAC, and the passenger and driver were male. The high prevalence of fatalities in these high-risk situations suggests that young driver-passenger dynamics may contribute to alcohol-related fatalities.

Practical Applications:

To curb RWI fatalities among underage passengers, countermeasures should focus not only on underage drinking drivers and riders, but also on drinking drivers of all ages. Prevention should increase focus on situations in which both the young passenger and young driver are males.

Keywords: Alcohol-related crashes, riding with an impaired driver, adolescents, passengers 5–20 years old

Introduction

In the United States, graduated driver licensing (GDL) restrictions have been effective tools to reduce youth involvement in crash fatalities (Vanlaar, Mayhew, et al., 2009; Fell, Jones, et al. 2011). Nevertheless, motor-vehicle crashes are the leading cause of unintentional injury death for every age 5 to 23 (Webb 2018), particularly when the driver is a young person driving at night (Chen, Baker, et al., 2000; Fell, Todd, et al., 2011; Shults & Williams, 2016). In 2015, 21% of all 15–20 year old (y/o) fatally-injured drivers had a BAC of 0.08 g/dL or higher (National Center for Statistics and Analysis, 2017) even though according to zero tolerance laws, the illegal limit for the underage group (<21 y/o) is between 0.00 g/dL and 0.02 g/dL, depending on the state. About 21% of the youth aged 20 y/o reported riding with a driver impaired by alcohol (RWI) in the past year (Li, Ochoa, et al., 2018). Examining a survey of high school students in Canada and the United States, Leadbeater and colleagues (Leadbeater, Foran, et al., 2008) reported that 52%–55% of the students reported “ever” riding with an impaired driver (RWI) aged 21 y/o or more, while 21%–33% of the students reported “ever” riding with an impaired peer (Leadbeater, Foran, et al., 2008). There is consensus that RWI constitutes a major public health concern as RWI is not only a major health-risking behavior, but also known as an antecedent of future driving while impaired (DWI) by alcohol (NCSA, 2012; Evans-Whipp, Plenty, et al., 2013; Li, Simons-Morton, et al., 2014).

RWI among teenagers has been found to be associated with rural residence (O’Malley & Johnston, 1999), and increasing with age (Sabel, Bensley, et al., 2004). The association of RWI with sex or race/ethnicity is less clear. Some studies have found that young females are more likely to RWI than males (Jelalian, Alday, et al., 2000; Harris, Johnson, et al., 2017). One study reported that those most likely to RWI were males (Everett, Shults, et al., 2001). Still, other studies found no significant association between passenger’s sex and RWI (Adlaf, Mann, et al., 2003; Hultgren, Turrisi, et al., 2018). One study by Grube and Voas found no association between the passengers’ race/ethnicity and their likelihood of RWI (Grube & Voas, 1996). Yet other studies found RWI more common among Latino youth than non-Latino White youth (O’Malley & Johnston, 1999; Walker, Treno, et al., 2003; Yellman, Bryan, et al., 2020). Vaca and colleagues also found RWI to be common among Latino youth, but only at certain ages (Vaca, Li, et al., 2016).

This study aimed to characterize young passengers who did RWI. More specifically, we aimed to assess the associations of sex and race/ethnicity with crashes in which passengers aged 15–20 y/o were killed while riding with a peer driver also aged 15–20 y/o. We hypothesized that most young passenger fatalities in the context of RWI occur on weekend nights because alcohol use and DWI among young drivers are prevalent on weekend nights (Tin, Ameratunga, et al., 2008; Goncy & Mrug, 2013). We examined whether the percentage of fatally injured passengers aged 15–20 y/o who died while riding with a peer driver was higher for Latinos than for any other racial/ethnic group. Because of sex difference in DWI involvement (Romano, Kelley-Baker, et al., 2008; Vaca, Romano, et al., 2014; Webster, Staton, et al., 2019), we expected to confirm that passengers aged 15–20 y/o are more likely to die in a RWI crash while riding with a male peer than a female peer. We also assessed whether the sexes of both the driver and passenger moderates that effect. We hypothesized that among passengers aged 15–20 y/o who died while riding with a peer also aged 15–20 y/o, the likelihood who were engaged in RWI at the time of the crash was higher when the driver was a young male driving a young male passenger (Mp-Md) than when a young male was driving a young female passenger (Fp-Md).

Methods

Data

Crash data were obtained from the 2010–2018 Fatality Analysis Reporting System (FARS). After discarding fatalities that involved vehicles other than passenger cars (e.g., buses, snowmobiles, motorcycles, trucks), crashes with missing information on drivers’ age, and crashes outside the scope of this study (e.g., police chases; non-moving vehicles), 5,673 fatally-injured passengers aged 15–20 y/o remained in the file. When assessing drivers’ alcohol use and to avoid double counting, drivers of vehicles in cases in which more than one passenger was present at the time of the crash were counted only once. Of the 5,673 fatally injured passengers aged 15–20 y/o in the file, of particular interest were the 3,542 who were riding with a driver also aged 15–20 y/o at the time of the crash.

Measures

Blood Alcohol Concentration (BAC).

About 65% of all drivers in the file have a measured BAC. Using multiple imputation, the FARS estimates the BAC of those with a missing BAC measure (Subramanian, National Center for, et al., 2002). We grouped drivers in three BAC categories: BAC=0.00 g/dL; 0.00 /dL < BAC < 0.08 g/dL; and BAC ≥ 0.08 g/dL.

Day of the week and time of the day:

Crashes were grouped as occurring on weekends (Friday to Sunday), or weekdays (remaining days), and either at nighttime (from 8 p.m. to 6 a.m.) or daytime (remaining hours).

Number of occupants:

Vehicles were grouped as carrying two versus more than two occupants (the driver and the 15–20 y/o passenger(s)) at the time of the crash.

Sex of passenger & driver:

For cases in which the fatally injured passenger was the sole passenger of the car, we considered whether the fatally injured passenger was a female riding with a male driver (Fp-Md), a female riding with a female driver (Fp-Fd), a male passenger riding with a male driver (Mp-Md), or a male passenger riding with a female driver (Mp-Fd). A fifth level was added to indicate when there were three or more occupants.

Rural vs. Urban setting.

Using FARS coding we assigned each crash to either a rural or urban setting.

Race and ethnicity.

Since 1988, the National Highway Traffic Safety Administration (NHTSA), working with the National Center for Health Statistics (NCHS), has been matching the records of road users fatally injured in crashes with their death certificate information in the NCHS Hyde cause-of-death (HCOD) file. This information appears in the FARS, although only on the deceased (i.e., the race and ethnicity of the surviving drivers is missing). The FARS informs separately on the deceased’s race (variable “Race,” with 19 categories, including White, Black, American Indian, Other, and unknown) and ethnicity (variable “Hispanic Origin,” with 9 categories including Mexican, Puerto Rican, Cuban, Other Hispanic Origin, and Unknown). For this study, the following four groups were considered: Latinos, non-Latino Blacks, non-Latino Whites, and non-Latino of Other race.

Statistical Analyses

We conducted cross-tables to examine the bivariate distribution of demographics and crash characteristics related to riding with a drinking driver. For each bivariate condition, prevalence of drivers at each of the three BAC level under examination were estimated. Comparisons were based on the 95% confidence intervals (95%CI) of the prevalence estimates Next, we ran a multinomial logistic regression model to assess the joint contribution of all factors identified by the bivariate analyses as contributors to the likelihood that fatally-injured adolescent passengers age 15–20 y/o were riding with a peer-aged driver with a 0.00g/dL<BAC<0.08 g/dL, and BAC≥0.08g/dL than with respect to BAC=0.00 g/dL (the reference level of the dependent variable). Main effects as well as dual interactions between all main effects were examined. We used SAS v9.4 for all analyses. We accounted for the additional variance introduced by the multiple imputation of BAC values by: (1) running 10 separate regressions, one for each of the 10 imputed BAC values; and (2) summarizing the results while accounting for standard errors with the Proc MIanalyze SAS procedure. Analyses were conducted between October 2019 and December 2020.

Results

Table 1 shows that between 2010 and 2018, a total of 5,673 passengers aged 15–20 y/o were killed while riding in passenger cars with a driver of known age. Of the 5,673 passengers aged 15–20 y/o in the file, a total of 3,542 (62.4%) died while riding with a driver also aged 15–20 y/o; a percentage significantly larger than the 18.7%, 7.0%, and 11.8% who died while riding with a driver aged 21–25 y/o, 26–35 y/o, and 36 y/o and over, respectively. Furthermore, Table 1 shows that the percent of the fatally injured passengers 15–20 y/o that died when riding with a BAC>0.00 g/dL driver was significantly lower when the driver was also aged 15–20 y/o (26.85%) than when the driver was aged 21–25 y/o (43.9%) or aged 26–35 y/o (42.0%), and significantly higher than when the driver was aged 36 y/o or over (18.5%).

Table 1.

Fatally injured passengers aged 15–20 y/o by driver age and BAC

Drivers’ BAC (g/dL) Driver Age (years)
15–20 21–25 26–35 36 and over All
N % (95% CI) N % (95% CI) N % (95% CI) N % (95%CI) N % (95%CI)
0.00 2592 73.2 596 56.1 231 58.0 547 81.5 3966 69.9
71.7 74.6 53.1 59.1 53.2 59.1 78.6 84.5 68.7 71.1
0.01–0.049 111 3.1 44 4.1 15 3.8 14 2.1 184 3.2
2.6 3.7 3.0 5.4 3.0 5.4 0.9 3.0 2.8 3.7
0.05–0.079 107 3.0 42 4.0 18 4.5 9 1.3 176 3.1
2.3 3.4 2.9 5.3 2.9 5.3 0.5 2.2 2.6 3.5
≥0.08 732 20.7 380 35.8 134 35.8 101 15.1 1347 23.7
20.1 22.8 33.5 39.3 33.5 39.3 12.2 17.6 23.2 25.4
BAC>0.00 950 26.8 466 43.9 167 43.9 124 18.5 1707 30.1
26.0 28.9 41.8 47.8 41.8 47.8 15.3 21.1 29.4 31.8
All (ROW) 3542 62.4 1062 18.7 398 7.0 671 11.8 5673 100.0
61.9 64.3 17.5 19.4 6.1 7.3 10.7 12.3
Open in a new tab

Source: FARS 2010–2018. BAC stands for blood alcohol concentration in grams per deciliter. BAC was either measured or imputed in the file. The values in row labeled All BAC>0.00 represents the sum of the previous 3 rows. 95% CI indicates 95% confidence interval. Total number of passengers aged 15–20 y/o do not sum up to 5,673 due to missing information on drivers’ age. The association between BAC level and driver’s age was statistically significant (p<.0001). Cells in gray were left empty due to small sample size.

Of the 950 fatalities of passengers aged 15–20 y/o who were riding with a BAC>0.00 g/dL driver also aged 15–20 y/o, 732 (77.1) was BAC≥0.08g/dL. Also shown in Table 1 is that for the 15–20 y/o passengers, riding with a BAC>0.00 g/dL driver was less prevalent when their drivers were 36 y/o and over than when of younger age. This result, at least in part, relates to many of the 36 y/o and over being adult family members, caretakers, or other non-peer adults of the 15–20 y/o passengers.

Factors contributing to individuals aged 15–20 y/o riding with BAC>0.00 g/dL drivers also aged 15–20 y/o.

The results in Table 2 further illustrates that most of the 15–20 y/o passengers that died when riding with a drinking peer, occurred when the underage peer driver was at BAC≥0.08g/dL. Table 2 also shows that the proportion of passengers aged 15–20 y/o who died while riding with a drinking peer did not vary significantly by the drivers’ race/ethnicity. The results in Table 2 also show that the percentage of passengers aged 15–20 y/o who died while riding with a drinking peer was significantly higher when the driver was a male than a female, both when the driver was 0.00 g/dL<BAC<0.08 g/dL (6.8% when the driver was male, 4.6% when female) and BAC≥0.08g/dL (22.9% when the driver was male, 14.9% when female). When there were only two occupants in the vehicle, both the sex of the passenger and the driver were associated with the driver being BAC≥0.08g/dL at the time of the crash. Although male drivers were more likely to be BAC≥0.08g/dL than female drivers, this prevalence is less common if the male was driving a female (Fp-Md, 19.1%) than another male (Mp-Md, 25.0%), although this difference was not statistically significant. Nevertheless, the degree of overlap between the confidence intervals was minimal and suggests the lack of significance could be attributed in part to sample size limitations. In the 59.6% of the cases in which there were more than two occupants in the crashed vehicle, the distribution of the drivers’ BAC did not differ statistically from cases in which there were only two occupants in the vehicle and the driver was a male. The urbanicity of the location of the crash was not significantly associated with the BAC of the driver. The percentage of BAC≥0.08g/dL drivers increased with age, a result that is expected since alcohol use increases with age (Masten, Faden, et al., 2009).

Table 2.

Percent of passengers aged 15–20 y/o who died while RWI drivers also aged 15–20 y/o by crash and driver’s characteristics and drivers’ BAC.

Driver’s BAC
BAC=0.00 0.00<BAC<0.08 BAC>=0.08
N (Col %) Row % 95%CI Row % 95%CI Row % 95%CI
Driver’s Race/Ethnicity Black 114 73.8 6.0 20.3
12.4 65.7 81.8 1.6 10.3 12.9 27.6
Latino 147 66.6 5.3 28.1
16.0 59.0 74.2 1.7 8.9 20.8 35.4
White 516 71.0 6.7 22.3
56.0 67.1 74.9 4.5 8.9 18.7 25.9
Other 144 67.4 5.9 26.7
15.6 59.7 75.0 2.1 9.7 19.5 34.0
Driver’s sex Male 2,539 70.3 6.8 22.9
71.7 68.5 72.1 5.8 7.7 21.3 24.6
Female 1,002 80.5 4.6 14.9
28.3 78.0 82.9 3.3 5.9 12.7 17.1
Sex of Passenger
& Driver 		Fp-Fd 248 83.5 2.6 13.9
7.0 78.9 88.1 0.6 4.6 9.6 18.2
Mp-Fd 147 85.1 2.4 12.5
4.2 79.4 90.8 0.1 4.8 7.2 17.9
F p -M d 359 76.0 4.9 19.1
10.1 71.6 80.4 2.7 7.2 15.0 23.1
M p -M d 677 69.2 5.8 25.0
19.1 65.7 72.7 4.1 7.6 21.7 28.2
3+ Occupant 2110 71.9 7.1 20.9
59.6 70.0 73.9 6.0 8.2 19.2 22.7
Urban/Rural Rural 1,089 73.3 6.1 20.5
30.8 70.7 76.0 4.7 7.6 18.1 22.9
Urban 2,438 73.2 6.1 20.7
68.9 71.4 74.9 5.2 7.1 19.1 22.3
Driver’s Age 15 89 82.9 2.8 14.3
2.5 75.1 90.7 0.1 6.1 7.0 21.5
16 425 85.6 3.2 11.1
12.0 82.3 89.0 1.6 4.9 8.1 14.1
17 700 79.9 5.1 15.0
19.8 76.9 82.9 3.4 6.7 12.4 17.7
18 929 72.4 6.5 21.2
26.2 69.5 75.2 4.9 8.0 18.6 23.8
19 774 67.4 8.0 24.6
21.9 64.0 70.7 6.1 9.9 21.6 27.7
20 624 64.3 7.0 28.7
17.6 60.5 68.0 5.0 9.0 25.2 32.3
Weekday/Weekend and Time of the Day WEEKDAY DAY 842 85.1 3.3 11.6
21.3 78.3 91.9 0.2 6.7 5.5 17.7
WEEKDAY NIGHT 703 56.4 7.6 35.9
17.8 48.6 64.3 3.4 11.8 28.3 43.5
WEEKEND DAY 796 84.3 2.3 13.5
20.1 78.3 90.3 0.0 4.7 7.9 19.1
WEEKEND NIGHT 1200 47.2 11.3 41.5
30.4 40.7 53.7 7.2 15.5 35.0 47.9
All 3541 73.2 6.1 20.7
100.0 71.7 74.6 5.4 6.9 19.3 22.0
Open in a new tab

Source: FARS 2010–2018. RWI indicates the percentage (%, and its 95% lower and upper confidence limits) of passengers aged 15–20 y/o who died while driving with a driver aged 15–20 y/o who recorded a positive blood alcohol concentration (BAC>0.00 g/dL). b Race/ethnicity is present in the FARS only on the deceased. Therefore, the race/ethnicity of the surviving drivers is missing. Members to more than one racial/ethnic group are included in the “Other” category. The first letter of the Fp-Fd, Mp-Fd, Fp-Md, and Mp-Md combinations indicates the driver’s sex, the second letter indicates the sex of the passenger. For instance, the Fp-Md combination indicates a Female driver and a Female passenger. “Weekend, day” denotes a crash that occurred on a Friday, Saturday, or Sunday Day. “Weekend, night” denotes a crash that occurred on a Friday, Saturday, or Sunday Night; “Weekday, day,” denotes a crash that occurred on a Monday, Tuesday, Wednesday, or Thursday Day. “Weekday, night” denotes a crash that occurred on a Monday, Tuesday, Wednesday, or Thursday Night. The number of occupants includes the driver.

As expected, the percentage of passengers aged 15–20 y/o who were riding with a BAC ≥0.08 g/dL peer was significantly higher on weekend nights (41.5%) or on weekdays at nighttime (35.9%) than on weekdays at daytime (11.6%) or weekends at daytime (13.5%). These results are consistent with current knowledge showing that drinking and driving is more prevalent at night, particularly on weekends (Romano, Kelley-Baker, et al., 2008).

Logistic Regression

Table 3 shows the odds ratio (OR) for the main effects included in the logistic regression modeling the BAC level of a 15–20 y/o individual driver of a fatally injured 15–20 y/o passenger. Overall, the results of the logistic regressions support the findings of the bivariate analyses. The fatally injured 15–20 y/o passengers were more likely to be found riding with a BAC ≥0.08 g/dL or a 0.00 g/dL <BAC<0.08 g/dL peer on a weekend night (OR=8.20, OR=6.20, respectively) or on a weekday night (OR=5.18; OR=3.90, respectively) than on a weekday at daytime. When there were only two occupants in the vehicle, the likelihood the driver was BAC ≥0.08 g/dL was significantly higher when both the driver and passenger were male than when both were female (OR=1.77). Although in Table 3, the overlapping confidence intervals corresponding to each level of the “Sex of the Passenger and Driver” variable seems to indicate that a vehicle with “3+ occupants” is as much likely to have been driven by a 0.00 g/dL < BAC < 0.08 g/dL, or BAC ⩾ 0.08 g/dL driver than a 2-occupants vehicle, such a lack of significance is caused by our partitioning of all 2+ occupant vehicles into 4 dyads (Mp-Fd, Fp-Md, Mp-Md, and Fp-Fd). After collapsing these four 2+ occupant dyads into a single level indicating there were only “2 occupants” in the vehicle, a comparison between this level and the “3+ occupants” level (not shown in Table 3) showed that the likelihood the peer driver was 0.00 g/dL < BAC <0.08 g/dL was significantly higher when there were 3+ occupants in the vehicle than when there were only 2 (OR=1.5, not shown in Table 3). The ORs the driver was 0.00 g/dL < BAC <0.08 g/dL or BAC ≥ 0.08 g/dL also increased with the driver’s age.

Table 3.

Multinomial logistic regression for variables modeling the likelihood of RWI by drivers’ BAC levels

Drivers’ BAC (Ref: BAC=0.00)
0.00<BAC<0.08 BAC≥0.08
OR 95%LCI 95%UCI OR 95%LCI 95%UCI
Day of the week and Time of the Day Weekend, day 1.03 0.69 1.55 1.13 0.56 2.25
Weekend, night 6.20 4.61 8.34 8.20 4.56 14.77
Weekday, night 3.90 2.77 5.49 5.18 2.69 9.96
Weekday, day (Ref)
Sex of Passenger & Driver Mp-Fd 0.77 0.17 3.57 0.76 0.36 1.57
Fp-Md 1.81 0.61 5.33 1.16 0.64 2.10
Mp-Md 2.17 0.80 5.89 1.77 1.04 3.03
3+ occupants 2.60 0.98 6.86 1.34 0.81 2.21
Fp-Fd (Ref)
Driver’s Age 15 y/o 0.96 0.18 5.22 1.30 0.60 2.82
17 y/o 1.74 0.86 3.54 1.45 0.88 2.38
18 y/o 2.13 0.98 4.63 2.04 1.25 3.32
19 y/o 2.75 1.25 6.04 2.46 1.56 3.88
20 y/o 2.44 1.07 5.57 3.24 2.01 5.22
16 y/ o (Ref)
Urbanicity Urban 0.93 0.67 1.29 0.97 0.78 1.22
Rural (Ref)
Race / Ethnicity Black 0.66 0.42 1.03 0.71 0.49 1.04
Latinx 1.07 0.67 1.69 0.99 0.74 1.32
Other 1.12 0.77 1.63 0.97 0.75 1.26
White (ref)
Open in a new tab

Source: FARS 2010–2018. OR stands for odds ratio. BAC stands for blood alcohol concentration in g/dL (grams per deciliter). The dependent variable (BAC) has 3 levels: BAC≥0.08 g/dL, 0.00 g/dL<BAC<0.08 g/dL, and BAC=0.00 g/dL, the reference group. BAC was either measured or imputed in the file. (Ref) indicates the reference level. The first letter of the Fp-Fd, Mp-Fd, Fp-Md, and Mp-Md combinations indicates, for crashes in which there were only 2 occupants in the vehicle, the passenger’s sex, the second letter indicates the sex of the driver. For instance, the Fp-Fd combination indicates a Female passenger riding with a Female driver. “Weekend, day” denotes a crash that occurred on a Friday, Saturday, or Sunday Day. “Weekend, night” denotes a crash that occurred on a Friday, Saturday, or Sunday Night; “Weekday, day,” denotes a crash that occurred on a Monday, Tuesday, Wednesday, or Thursday Day. “Weekday, night” denotes a crash that occurred on a Monday, Tuesday, Wednesday, or Thursday Night. The number of occupants includes the driver.

Discussion

Although studies focusing on youth involvement in alcohol-related fatal crashes are not new, most have focused on the fatally injured driver (Tefft, Williams, et al., 2013; Simons-Morton, Ehsani, et al., 2017). Studies focusing on the passengers who died while riding with an impaired driver (RWI) are far less frequent, and typically based on self-reported data (Poulin, Boudreau, et al., 2007; Cartwright & Asbridge, 2011). RWI studies based on crash data are less frequent in part due to the challenges posited by the absence of information on driving exposure (i.e., all events in the file are crashes), and the imprecise determination of drivers’ impairment. In this study, we attempted to address some of these limitations by looking at the BAC of the young passengers’ drivers, as BAC relates with impairment and RWI. Although BAC cannot accurately indicate impairment and subsequently any identification of RWI based on BAC lacks precision, we argue that passengers who were riding with a BAC ≥ 0.08 g/dL driver aged 15–20 y/o were RWI. Furthermore, because for underage, novice drivers, alcohol impairment may start at relatively low BACs (Peck, Gebers, et al., 2008), we argue that 15–20 y/o passengers were RWI when riding with drivers at any positive BAC. Regardless of the merits of considering 15–20 y/o drivers at any BAC > 0.00g/dL level as impaired, our finding that most (77.1%) of the BAC > 0.00 g/dL young drivers in the file were BAC ≥ 0.08 these criteria yield similar results. Such heavy drinking among the underage drivers not only occurred despite minimum legal drinking laws, but often occurred at nighttime, particularly on weekends, which suggests the driving of these minors related to some festive environments.

Our findings that most (62.4%) of the fatally-injured 15–20 y/o passengers died while riding with a driver also aged 15–20 y/o confirm previous reports showing that 57% of the teen passengers who died in a crash in 2018 were driven by another teenager (IIHS 2019); and shows that despite the large majority (about 89%) of licensed U.S. drivers being aged 21 y/o or older, most 15–20 y/o passengers who were fatally injured in a crash, died while riding with a peer. However, when alcohol is considered, we found that when a passenger 15–20 y/o died while riding with a drinking driver, it was less likely that the drinking driver was also a peer aged 15–20 y/o, than an older driver. This finding is in line with previous self-reports showing that 52%–55% of high school students self-reported “ever” RWI with a driver aged 21 y/o or more, and 21%–33% self-reported “ever” RWI with a peer (Leadbeater, Foran, et al., 2008). The finding that the drivers of 15–20 y/o passengers are more likely to be impaired when they are age 21 y/o or older suggests that zero-tolerance laws alone are not enough to prevent the death of passengers aged 15–20 y/o who die in an alcohol-related crash. As such, this finding points out that in order to curb RWI fatalities among underage passengers, it is necessary to implement and/or enhance the countermeasures that have been proven to be effective against drinking drivers of all ages (e.g., sobriety checkpoints; Fell, Lacey, et al., 2004).

While an alcohol-related fatality among 15–20 y/o passengers is more likely to occur when the passenger is riding with an older driver, this should not be viewed as an indication that fatalities of 15–20 y/o passengers that occur when they are riding with a 15–20 y/o driver are of no or little importance. As already shown, 62.4% of all 15–20 y/o passengers who died in a crash were riding with a 15–20 y/o driver. Thus, although for an individual 15–20 y/o passenger the likelihood that her/his driver is impaired is lower when the driver is also 15–20 y/o than when older, by sheer numbers, slightly more than half of the 15–20 y/o passengers who died while RWI died while riding with a 15–20 y/o driver (54% of BAC ≥ 0.08 g/dL drivers). This result emphasizes the need to also increase our efforts to implement and/or enhance the countermeasures that have been shown to be effective against underage drinking drivers (e.g., zero tolerance laws).

Our finding that for fatally injured 15–20 y/o passengers, the likelihood the driver was BAC ≥ 0.08 g/dL was lower when the driver was also 15–20 y/o than when the driver was older, and that most of these passengers that died were riding with another 15–20 y/o passenger, suggesting that while alcohol contributed to most fatalities when the driver was older than 21 y/o, reasons other than alcohol are behind a sizable number of fatalities involving 15–20 y/o drivers. Besides alcohol, distractions, inexperience, speeding, and drowsiness are some of the most frequent contributors to crashes among young and novice drivers (Groeger, 2006; Klauer, Guo, et al., 2014; Simons-Morton, Guo, et al., 2014). The finding that the likelihood of an RWI fatality when the driver was 0.00 g/dL < BAC < 0.08 g/dL was higher when more than one passenger was present at the time of the crash than when only one passenger was present may indicate that driving with multiple passengers is a source of additional distraction to the young drinking drivers, further increasing the odds that a passenger would die in a fatal crash with the driver at 0.00 g/dL < BAC < 0.08 g/dL. Regarding the lack of significance of this factor when the driver was BAC≥0.08g/dL, we speculate that when the driver is heavily impaired (BAC≥ 0.08g/dL), alcohol is the main source of risk and the distractions created by additional passengers do not contribute to crash risk as much as when the distraction occurs at lower BACs.

One of the aims of this study was to assess whether race/ethnicity was a factor contributing to RWI fatalities among 15–20 y/o passengers. In our analysis, we found that race/ethnicity was not a factor influencing the likelihood of RWI fatal crashes. Another study aim was to assess whether the sex of the 15–20 y/o drivers and passengers was associated with the passengers dying in alcohol-related crashes. We found that when there were only two occupants in the vehicle, the sexes of the young driver and young passenger affects the likelihood the passenger was RWI. Among the young passengers who died in an alcohol-related crash, it is less likely to find a young female passenger riding with a young male driver, than a male passenger riding with another young male. There is a need to better understand the context of young driver-passenger dynamics, particularly among young dyads, and how such dynamics affect alcohol use and alcohol-related crashes. It might be possible that variation in how the driver-passenger dynamics plays out could explain some of the discrepancies on the role of race/ethnicity on RWI reported in the literature. Regardless, advancing the understanding of the driver-passenger dynamics is needed for the design of efficient and effective interventions to deter young people from engaging in RWI.

Of course, the most effective form of prevention is reducing impaired driving by providing alternatives to driving and to drinking. With respect to driving, these alternatives include public transportation, riding sharing, parental responsibility legislation, and designated driver programs. Other evidence-based countermeasures include high-visibility enforcement of zero tolerance laws (Johnson, 2016); programs and interventions to reduce accessibility of alcohol to minors (Komro & Toomey, 2002; Flewelling, Grube, et al., 2013; Fell, Fisher, et al., 2009; Wagenaar, Harwood, et al., 2005); communities efforts to limit alcohol outlet density (Chen, Grube, et al., 2010); the enactment and enforcement of ordinances such as alcohol retailer compliance checks (Elder, Lawrence, et al., 2007; Erickson, Smolenski, et al., 2013); keg registration (Ringwalt & Paschall, 2011); or social host ordinances that include strict liability and civil penalties (Paschall, Lipperman-Kreda, et al., 2014).

The findings provide some support for extending extant GDL passenger restrictions to age 20. Older teen-young adult might be neurodevelopmentally mature enough to be able to navigate social situation or context that would allow them to avoid engaging in RWI or DWI. Also, it is possible that better trained novice drivers aged 18–20 y/o would be able to navigate the context that might typically precede an impairing situation associated with alcohol use (despite being illegal) with better chances to avoid a crash and survive than less skilled drivers. Although this study provides some support for extending GDL programs, the evidence is far from conclusive and needs more examination.

This study has several limitations. Impairment by alcohol cannot be precisely established from the FARS. Drugs other than alcohol may have also contributed to the crashes examined. Unfortunately, as indicated by the agency that manages the database, drug-related crashes cannot be reliably studied from the FARS (Berning & Smither, 2014; Romano, Torres-Saavedra, et al. 2017). Information on driver’s race and ethnicity was incomplete, as it is only available on the deceased occupants in FARS. Another important limitation of this study is that analyses are not adjusted by crash exposure. Although relevant and novel, our study was based only on fatal crashes, subsequently it does not take nonfatal and non-crashed RWI events into account.

Conclusions

Among 15–20 y/o passenger deaths, the majority occurred while riding with peer drivers. In nearly 27% of these cases, the driver had been drinking, and when the driver was drinking, about 77% of them had been drinking heavily. In order to curb RWI fatalities among underage passengers, it is necessary to enhance the implementation of countermeasures focused not only on underage drinking drivers, but also policies restricting drinking drivers of all ages.

Acknowledgments

This work was supported by the National Institute on Alcohol Abuse and Alcoholism (grant number R21AA026346).

Biography

Eduardo Romano, Ph.D., is a Senior Research Scientist at the Pacific Institute for Research and Evaluation. An economist by training, his research interests have focused on risk-related behaviors as well as on the analyses of risk-reducing and risk-managing policies. His recent interests include risk perceptions and impaired driving, particularly among children, adolescents, women, and minorities. He is conducting research on the effectiveness of alcohol interlock devices, and on the developmental trajectories towards impaired driving outcomes. Among others organizations, he is a member of the Research Society on Alcoholism, the Research Society on Marijuana, ICADTS, National Hispanic Science Network, and the TRB Committee on Impaired Driving

James C. Fell is currently a Principal Research Scientist with the National Opinion Research Center (NORC) at the University of Chicago in the Bethesda, Maryland office. Mr. Fell worked at the National Highway Traffic Safety Administration (NHTSA) from 1969 to 1999 and has over 50 years of traffic safety and alcohol policy research experience. He has authored or coauthored over 170 publications in book chapters, scientific journals and conference proceedings. He is a long-time member of the Association for the Advancement of Automotive Medicine (AAAM)(since 1969), including serving as president (1988), treasurer (1985–86); and elected as a Fellow (1994). He received the AAAM Award of Merit in 2019. Mr. Fell is currently President-Elect of the International Council on Alcohol, Drugs, and Traffic Safety (ICADTS) and the 2013 recipient of the ICADTS Widmark Award and the 2019 recipient of the ICADTS Haddon Award. In 2015, Mr. Fell received the James J. Howard Highway Safety Trailblazer Award from the Governors Highway Safety Association (GHSA) for sustained outstanding leadership in endeavors that significantly improve highway safety and the Kevin Quinlan Advocacy Award from the Maryland Highway Safety Office. He has both a Bachelor’s and Master’s degree in Human Factors Engineering from the State University of New York at Buffalo.

Kaigang Li, PhD, MEd, is an Assistant Professor in the Department of Health and Exercise Science, Colorado State University (CSU) and the Director of the Laboratory for Assessment and Promotion of Physical Activity and Health (APPAH) at CSU. One of his research focuses is adolescent health and risk behavior including teenage risky driving and alcohol/drug-impaired driving.

Bruce Simons-Morton, EdD, MPH. Until 2019 Bruce Simons-Morton was Senior Investigator at the National Institute of Child Health and Human Development where he directed an intramural program of research on adolescent and young adult health behavior that included studies on substance use and driving. He is currently Senior Research Scientist and Director of Impaired Driving Research, Evaluation, and Analyses at the Virginia Tech Transportation Institute.

Federico E. Vaca, MD, MPH is a Professor and Vice Chair in the Yale School of Medicine’s Department of Emergency Medicine. He holds a secondary appointment in the Yale Child Study Center. Dr. Vaca is the founding Director of the Yale Developmental Neurocognitive Driving Simulation Research Center (DrivSim Lab). As an NIH funded physician-scientist, his research has focused on the use of quantitative, qualitative, and mixed methods approaches to understand developmental, behavioral, and socioecological relationships associated with racial/ethnic disparities in motor vehicle crash related morbidity and mortality, risky driving behavior in youth, and alcohol use disorders. His NIH funded research includes study of Latino and non-Latino populations in the context of alcohol/substance use disorders, alcohol/drugged driving, riding with alcohol/drug impaired drivers, and prevention-related policies. He has been a Collaborating-Visiting Scholar at the NIH’s Eunice Kennedy Shriver National Institute of Child Health & Human Development, Division of Intramural Population Health Research, Health Behavior Branch. Dr. Vaca has Chaired and served on NIH scientific grant review panels. He currently serves on standing committees for the National Academies of Sciences, Engineering, and Medicine’s Transportation Research Board with committee focus on vehicle operator education and regulation as well as simulation and measurement of vehicle and operator performance. After being appointed by the U.S. Secretary of Health and Human Services, he recently completed his service on the Board of Scientific Counselors for the CDC’s National Center for Injury Prevention and Control. In April 2019, Dr. Vaca was awarded the David J. Leffell Prize for Clinical Excellence, an annual award given to only one physician in the Yale School of Medicine who best demonstrates the highest level of clinical expertise, commitment to teaching, and the highest standards of care and compassion for patients. As noted by the Yale School of Medicine, “Dr. Vaca was awarded the 2019 David J. Leffell Prize for Clinical Excellence in recognition of his extraordinary contributions to the Yale School of Medicine (YSM) as a clinician, educator, and researcher who continues to break new ground in the area of traumatic injury prevention.”

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adlaf EM, Mann RE and Paglia A (2003). “Drinking, cannabis use and driving among Ontario students.” CMAJ : Canadian Medical Association journal = journal de l’Association medicale canadienne 168(5): 565–566. [PMC free article] [PubMed] [Google Scholar]
  2. Berning A and Smither DD (2014). Understanding the limitations of drug test information, reporting, and testing practices in fatal crashes. (Traffic Safety Facts Research Note. DOT HS 812 072) Washington, DC, National Highway Traffic Safety Administration. [Google Scholar]
  3. Cartwright J and Asbridge M (2011). “Passengers’ decisions to ride with a driver under the influence of either alcohol or cannabis.” Journal of studies on alcohol and drugs 72(1): 86–95. [DOI] [PubMed] [Google Scholar]
  4. Chen LH, Baker SP, Braver ER and Li G (2000). “Carrying passengers as a risk factor for crashes fatal to 16- and 17-year-old drivers.” JAMA 283(12): 1578–1582. [DOI] [PubMed] [Google Scholar]
  5. Chen MJ, Grube JW and Gruenewald PJ (2010). “Community alcohol outlet density and underage drinking.” Addiction 105(2): 270–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Elder RW, Lawrence BA, Janes G, Brewer RD, Toomey TL, Hingson RW, Naimi TS, Wing SG and Fielding J (2007). “Enhanced enforcement of laws prohibiting sale of alcohol to minors: systematic review of effectiveness for reducing sales and underage drinking.” Transportation research circular 2007(E-C123): 181–188. [Google Scholar]
  7. Erickson DJ, Smolenski DJ, Toomey TL, Carlin BP and Wagenaar AC (2013). “Do alcohol compliance checks decrease underage sales at neighboring establishments?” Journal of studies on alcohol and drugs 74(6): 852–858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Evans-Whipp TJ, Plenty SM, Toumbourou JW, Olsson C, Rowland B and Hemphill SA (2013). “Adolescent exposure to drink driving as a predictor of young adults’ drink driving.” Accident Analysis & Prevention 51: 185–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Everett SA, Shults RA, Barrios LC, Sacks JJ, Lowry R and Oeltmann J (2001). “Trends and subgroup differences in transportation-related injury risk and safety behaviors among high school students, 1991–1997.” Journal of Adolescent Health 28(3): 228–234. [DOI] [PubMed] [Google Scholar]
  10. Fell JC, Fisher DA, Voas RB, Blackman K and Tippetts AS (2009). “The impact of underage drinking laws on alcohol‐related fatal crashes of young drivers.” Alcoholism: Clinical and Experimental Research 33(7): 1208–1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fell JC, Jones K, Romano E and Voas R (2011). “An evaluation of graduated driver licensing effects on fatal crash involvements of young drivers in the United States.” Traffic injury prevention 12(5): 423–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fell JC, Lacey JH and Voas RB (2004). “Sobriety checkpoints: evidence of effectiveness is strong, but use is limited.” Traffic injury prevention 5(3): 220–227. [DOI] [PubMed] [Google Scholar]
  13. Fell JC, Todd M and Voas RB (2011). “A national evaluation of the nighttime and passenger restriction components of graduated driver licensing.” Journal of Safety Research 42(4): 283–290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Flewelling RL, Grube JW, Paschall MJ, Biglan A, Kraft A, Black C, Hanley SM, Ringwalt C, Wiesen C and Ruscoe J (2013). “Reducing youth access to alcohol: Findings from a community-based randomized trial.” American journal of community psychology 51(1–2): 264–277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Goncy EA and Mrug S (2013). “Where and when adolescents use tobacco, alcohol, and marijuana: comparisons by age, gender, and race.” Journal of studies on alcohol and drugs 74(2): 288–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Groeger JA (2006). “Youthfulness, inexperience, and sleep loss: the problems young drivers face and those they pose for us.” Injury prevention 12(suppl 1): i19–i24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Grube JW and Voas RB (1996). “Predicting underage drinking and driving behaviors.” Addiction (Abingdon, England) 91(12): 1843–1857. [DOI] [PubMed] [Google Scholar]
  18. Harris SK, Johnson JK, Sherritt L, Copelas S, Rappo MA and Wilson CR (2017). “Putting Adolescents at Risk: Riding With Drinking Drivers Who Are Adults in the Home.” Journal of Studies on Alcohol and Drugs 78(1): 146–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hultgren BA, Turrisi R, Mallett KA, Ackerman S, Larimer ME, McCarthy D and Romano E (2018). “A Longitudinal Examination of Decisions to Ride and Decline Rides with Drinking Drivers.” Alcoholism, clinical and experimental research. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. IIHS (2019). “Teenagers.” Insurance Institute for Highway Safety, Fatality Facts 2018 https://www.iihs.org/topics/fatality-statistics/detail/teenagers(Accesed Septemer 20, 2020).
  21. Jelalian E, Alday S, Spirito A, Rasile D, Nobile CRIH and P. R. I. U. S. A. Brown University School of Medicine (2000). “Adolescent motor vehicle crashes: the relationship between behavioral factors and self-reported injury.” Journal of Adolescent Health 27(2): 84–93. [DOI] [PubMed] [Google Scholar]
  22. Johnson MB (2016). “A successful high‐visibility enforcement intervention targeting underage drinking drivers.” Addiction 111(7): 1196–1202. [DOI] [PubMed] [Google Scholar]
  23. Klauer SG, Guo F, Simons-Morton BG, Ouimet MC, Lee SE and Dingus TA (2014). “Distracted driving and risk of road crashes among novice and experienced drivers.” New England journal of medicine 370(1): 54–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Komro KA and Toomey TL (2002). “Strategies to prevent underage drinking.” Alcohol Research & Health 26(1): 5. [PMC free article] [PubMed] [Google Scholar]
  25. Leadbeater BJ, Foran K and Grove‐White A (2008). “How much can you drink before driving? The influence of riding with impaired adults and peers on the driving behaviors of urban and rural youth.” Addiction 103(4): 629–637. [DOI] [PubMed] [Google Scholar]
  26. Li K, Ochoa E, Vaca FE and Simons-Morton B (2018). “Emerging adults riding with marijuana-, alcohol-, or illicit drug–impaired peer and older drivers.” Journal of Studies on Alcohol and Drugs 79(2): 277–285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li K, Simons-Morton BG, Vaca FE and Hingson R (2014). “Association between riding with an impaired driver and driving while impaired.” Pediatrics 133(4): 620–626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Masten AS, Faden VB, Zucker RA and Spear LP (2009). “A developmental perspective on underage alcohol use.” Alcohol research & health : the journal of the National Institute on Alcohol Abuse and Alcoholism 32(1): 3–15. [PMC free article] [PubMed] [Google Scholar]
  29. National Center for Statistics and Analysis (2017). Young drivers: 2015 data. (Traffic Safety Facts. Report No. DOT HS 812 363) Washington, DC, National Highway Traffic Safety Administration. [Google Scholar]
  30. NCSA (2012). TRAFFIC SAFETY FACTS 2012. TRAFFIC SAFETY FACTS. N. H. T. S. Administration. Washington, DC 20590 [Google Scholar]
  31. O’Malley PM and Johnston LD (1999). “Erratum - In: Drinking and Driving Among US High School Seniors.” American journal of public health 89(9): 1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Paschall MJ, Lipperman-Kreda S, Grube JW and Thomas S (2014). “Relationships between social host laws and underage drinking: Findings from a study of 50 California cities.” Journal of studies on alcohol and drugs 75(6): 901–907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Peck RC, Gebers MA, Voas RB and Romano E (2008). “The relationship between blood alcohol concentration (BAC), age, and crash risk.” Journal of safety research 39(3): 311–319. [DOI] [PubMed] [Google Scholar]
  34. Poulin C, Boudreau B and Asbridge M (2007). “Adolescent passengers of drunk drivers: a multi-level exploration into the inequities of risk and safety.” Addiction 102(1): 51–61. [DOI] [PubMed] [Google Scholar]
  35. Ringwalt CL and Paschall MJ (2011). “The utility of keg registration laws: A cross-sectional study.” Journal of Adolescent Health 48(1): 106–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Romano E, Kelley-Baker T and Voas RB (2008). “Female involvement in fatal crashes: increasingly riskier or increasingly exposed?” Accident; analysis and prevention 40(5): 1781–1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Romano E, Torres-Saavedra P, Voas RB and Lacey JH (2017). “Marijuana and the risk of fatal car crashes: What can we learn from FARS and NRS data?” Journal of Primary Prevention 38(3): 315–328. [DOI] [PubMed] [Google Scholar]
  38. Sabel JC, Bensley LS and Van Eenwyk J (2004). “Associations between adolescent drinking and driving involvement and self-reported risk and protective factors in students in public schools in Washington State.” Journal of studies on alcohol 65(2): 213–216. [DOI] [PubMed] [Google Scholar]
  39. Shults RA and Williams AF (2016). “Graduated driver licensing night driving restrictions and drivers aged 16 or 17 years involved in fatal night crashes—United States, 2009–2014.” Morbidity and Mortality Weekly Report 65(29): 725–730. [DOI] [PubMed] [Google Scholar]
  40. Simons-Morton B,G, J. Ehsani P, Gershon P, S. Klauer G and T. Dingus A (2017) “Teen Driving Risk and Prevention: Naturalistic Driving Research Contributions and Challenges.” Safety 3 DOI: 10.3390/safety3040029. [DOI] [Google Scholar]
  41. Simons-Morton BG, Guo F, Klauer SG, Ehsani JP and Pradhan AK (2014). “Keep your eyes on the road: Young driver crash risk increases according to duration of distraction.” Journal of Adolescent Health 54(5): S61–S67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Subramanian R,S National Center for and Analysis (2002). Transitioning to multiple imputation : a new method to impute missing blood alcohol concentration (BAC) values in FARS. Washington, D.C, National Center for Statistics and Analysis, Research and Development. [Google Scholar]
  43. Tefft BC, Williams AF and Grabowski JG (2013). “Teen driver risk in relation to age and number of passengers, United States, 2007–2010.” Traffic injury prevention 14(3): 283–292. [DOI] [PubMed] [Google Scholar]
  44. Tin ST, Ameratunga S, Robinson E, Crengle S, Schaaf D and Watson P (2008). “Drink driving and the patterns and context of drinking among New Zealand adolescents.” Acta paediatrica (Oslo, Norway : 1992) 97(10): 1433–1437. [DOI] [PubMed] [Google Scholar]
  45. Vaca FE, Li K, Hingson R and Simons-Morton BG (2016). “Transitions in Riding With an Alcohol/Drug-Impaired Driver From Adolescence to Emerging Adulthood in the United States.” Journal of studies on alcohol and drugs 77(1): 77–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Vaca FE, Romano E and Fell JC (2014). “Female Drivers Increasingly Involved in Impaired Driving Crashes: Actions to Ameliorate the Risk.” Academic Emergency Medicine 21(12): 1485–1492. [DOI] [PubMed] [Google Scholar]
  47. Vanlaar W, Mayhew D, Marcoux K, Wets G, Brijs T and Shope J (2009). “An evaluation of graduated driver licensing programs in North America using a meta-analytic approach.” Accident; analysis and prevention 41(5): 1104–1111. [DOI] [PubMed] [Google Scholar]
  48. Wagenaar AC, Harwood EM, Silianoff C and Toomey TL (2005). “Measuring public policy: The case of beer keg registration laws.” Evaluation and Program Planning 28(4): 359–367. [Google Scholar]
  49. Walker S, Treno AJ, Grube JW and Light JM (2003). “Ethnic Differences in Driving After Drinking and Riding With Drinking Drivers Among Adolescents.” Alcoholism: Clinical and Experimental Research 27(8): 1299–1304. [DOI] [PubMed] [Google Scholar]
  50. Webb CN (2018). Motor vehicle traffic crashes as a leading cause of death in the United States, 2015. Traffic Safety Facts. Washington, DC, National Highway Traffic Safety Administration. [Google Scholar]
  51. Webster JM, Staton M and Dickson MF (2019). “Brief Report: Sex Differences in Substance Use, Mental Health, and Impaired Driving Among Rural DUI Offenders.” The American journal on addictions 28(5): 405–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Yellman MA, Bryan L, Sauber-Schatz EK and Brener N (2020). “Transportation Risk Behaviors Among High School Students - Youth Risk Behavior Survey, United States, 2019.” MMWR supplements 69(1): 77–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES