Skip to main content

Evaluating the effect of drunk driving on fatal injuries among vulnerable road users in Taiwan: a population-based study

Abstract

Background

Most studies have focused on injuries sustained by intoxicated drivers themselves, but few have examined the effect of drunk driving on injury outcomes among VRUs (vulnerable road users) in developing countries. This study aims to evaluate the effect of drunk driving on fatal injuries among VRUs (pedestrians, cyclists, or motorcyclists).

Methods

The data were extracted from the National Taiwan Traffic Crash Dataset from January 1, 2011, to December 31, 2019. Crashes involving one motorized vehicle and one VRU were considered. This study examines the effect of drunk driving by estimating multivariate logistic regression models of fatal injuries among VRUs after controlling for other variables.

Results

Among 1,416,168 casualties, the fatality rate of VRUs involved in drunk driving was higher than that of general road users (2.1% vs. 0.6%). Drunk driving was a significant risk factor for fatal injuries among VRUs. Other risk factors for fatal injuries among VRUs included VRU age ≥ 65 years (adjusted odds ratio [AOR]: 5.24, 95% confidence interval [CI]: 5.53–6.07), a nighttime accident (AOR: 4.52, 95% CI: 4.22–4.84), and being hit by a heavy-duty vehicle (AOR: 2.83, 95% CI: 2.26–3.55). Subgroup analyses revealed a linear relationship between driver blood alcohol concentration (BAC) and the risk of fatal injury among motorcyclists. Motorcyclists exhibited the highest fatality rate when they had a BAC ≤ 0.03% (AOR: 3.54, 95% CI: 3.08–4.08).

Conclusion

Drunk driving was associated with a higher risk of fatality for all VRUs. The risk of fatal injury among motorcyclists was linearly related to the BAC of the drunk drivers. Injuries were more severe for intoxicated motorcyclists, even those with BAC ≤ 0.03%, which is within the legal limit.

Peer Review reports

Background

Alcohol acts as a central nervous system depressant that alters the level of consciousness [1, 2] and reduces the attentional and behavioral control of drivers [3, 4]. Drunk driving is a risky behavior; drunk drivers may judge the traffic condition improperly because they exhibit overestimation of personal abilities [5], excessive bravery [6], and a tendency to be affected by false memory [7]. Alcohol also interferes with visual acuity, perception, and psychomotor function; reduces reactions to impulses and environmental vigilance; and impairs the postural control of drivers [8,9,10,11,12,13]. Moreover, impaired decision-making [14] and information processing are evident among drivers with a positive blood alcohol concentration (BAC) [15]. Simulation studies have also demonstrated negative effects of alcohol on driving speed [16, 17], accelerating and braking behavior [18], and lane positioning [19].

The positive correlation between drunk driving and motor vehicle crashes (MVCs) has been well documented [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34]. Even with a mildly elevated BAC (0.01–0.03%), drunk drivers cause more MVCs than do drivers who have not consumed alcohol [35,36,37]. Drunk driving not only increases the MVC risk but also results in more fatal crashes [28, 38, 39]. Zador et al. revealed that drivers with BACs < 0.1% contributed to more fatal injuries to both themselves and other road users [40]. Reynaud et al. analyzed the French police records through a 5-year period and found that 31.5% of those who died in an accident had a positive BAC, and 9.8% of them had a BAC over the legal limit [41]. The detrimental effect of alcohol use has also been confirmed in another French study [42], suggesting that fatigue, when combined with alcohol, presented a particularly high risk of crashes leading to death or serious injuries. Moreover, the victims of alcohol-impaired driving have higher risks of hospitalization, hypotension, Glasgow Coma Scale (GCS) scores, and events of cardiac arrest [28]. A retrospective analysis of 474 autopsy reports documenting fatalities in traffic crashes revealed 177 victims with a positive BAC [39]. Substance and alcohol use was also reported to be associated with reduced reaction times [43], as well as several risky behaviours such as driving without a seatbelt [44], unlicensed driving [45], and speeding [46].

In traffic accidents, vulnerable road users (VRUs)—motorcyclists, bicyclists, and pedestrians—sustain severe injuries and death at a higher rate than motorists. This is because without the protection afforded by a metal structure, VRUs generally sustain more severe injuries than car occupants [47]. Moreover, car drivers may have difficulty identifying or perceiving VRUs in traffic due to their being poorly visible and having a small size, which may increase the severity of a crash in the event of an accident [48,49,50,51].

Most studies have focused on injuries sustained by intoxicated drivers themselves, but few have examined the effect of drunk driving towards VRUs (vulnerable road users) in developing countries such as Taiwan. To fill this research gap, we analyzed Taiwan’s national police crash data and investigated the effects of drunk driving with other risk factors on fatal injuries among VRUs.

Methods

Study participants and data source

This study analyzed the National Taiwan Traffic Crash Dataset from January 1, 2011, to December 31, 2019. The dataset is administrated by the National Police Agency of Taiwan. Experienced police crash investigators are assigned to arrive at the scene and record the information which includes crash, vehicle, and victim files. Crash files contain data on road traffic crash characteristics such as time of crash, date of crash, weather condition, light condition, and various environmental factors (such as geographic location, speed limit, type and condition of road, and apparent distance). Vehicle files contain data on characteristics of the vehicle involved in the crash, such as first point of impact, type of vehicle, and vehicle maneuver. Furthermore, data on victim characteristics such as age, sex, injury severity level, license status, BAC, travel purpose, and restraint use are contained in the victim files. Similar to those in other countries, the Dataset is considered complete for multi-vehicle crashes but less complete for single-vehicle crashes; such an underreporting problem is less of a concern as the present study focuses on multi-vehicle crashes (i.e., an automobile and a VRU). In addition, variables that contain numerous missing data (e.g., hit-and-run crashes) or are unreliable (e.g., mobile phone use) were not considered in the current research. Every road traffic-related crash reported to the police was recorded in the dataset, which is maintained by the National Police Agency of Taiwan. In this study, we focused on crashes involving one automobile and one VRU (motorcyclist, cyclist, or pedestrian). Figure 1 illustrates the data extraction flowchart for this study. We excluded single-vehicle crashes, multiple-vehicle (> 2) crashes, VRU–VRU crashes, and crashes involving no VRUs from the dataset. Finally, we removed cases with missing data because we used a complete case analysis approach for our data analysis. This study was approved by the Joint Institutional Review Board of Taipei Medical University (number: N202007045). The current research analysed national crash data without individuals’ confidential information such as names or identity numbers. As a result, the Institutional Review Board affiliated with Taipei Medical University waived the informed consent.

Fig. 1
figure 1

Selection of casualties

Study variables

Two injury severity levels were recorded: fatal injury (death within 24 h after crash) and nonfatal injuries (sustained injuries and survived for > 24 h). We also collected basic demographic data such as age, sex, participant’s safety behaviors, including helmet use by motorcyclists and bicyclists, BAC level, and the license status of drivers and motorcyclists. Because bicyclists and pedestrians are not required to be tested for alcohol use in the event of traffic accident, their BAC levels were not included in the present analysis.

Temporal variables included in this research were the time of the crash (rush hour, daytime, night, or early morning) and whether the accident occurred on a weekday or weekend. Rush hour was defined as 07:00 AM to 08:59 AM and 5:00 PM to 7:59 PM, daytime was defined as 09:00 AM to 4:59 PM, evening was defined as 8:00 PM to 11:59 PM, and nighttime was defined as 12:00 AM to 06:59 AM.

The following road and environmental factors were analyzed: weather (fine weather refers to sunny and cloudy days; adverse weather includes rainy, snowy, foggy, or sandy conditions and strong winds), and light conditions (no light at night, illuminated at night, morning or dawn, and daytime with natural light; if the incident was in a tunnel or underpass, the setting was deemed night). Taiwan has six municipalities: Kaohsiung, New Taipei, Taichung, Tainan, Taipei, and Taoyuan. Other regions are defined as counties. Several road conditions were considered in this study, including road type (crossroad or not), road surface conditions (slippery road includes snowy/icy, oily, muddy, or damp road), road defect (intact road surface or a defective road, meaning soft terrain, uneven road, or road with pit or hole), and driver’s sightline (clear sight or obstacle in sight). Speed limit was divided into less than 50 km/h and ≥ 50 km/h. Injured body regions of VRUs were categorized as a head and neck injury and other injuries including the chest, abdomen, back, pelvis, and extremities. Table 1 illustrates variables included in analysis.

Table 1 Description of each variable

Statistical analysis

We first compared the distribution of fatal injuries by demographic factors, behaviors, vehicle attributes, crash characteristics, environmental factors, time factors, and crash types. A p value < 0.2 was used as the cutoff point to incorporate risk factors into multivariate analysis. Multiple logistic regression analysis with backward selection was used to calculate the adjusted odds ratios (AORs). Multicollinearity was assessed using Cramer’s V and the chi-square independent test. A subgroup analysis was conducted separately for motorcyclists, bicyclists, and pedestrians. A full model (automobile VRUs) was first estimated, followed by three additional models: an automobile–motorcycle (A-M) model, an automobile–bicycle (A-B) model, and an automobile–pedestrian (A-P) model. Statistical significance was defined as p < 0.05. The binary logistic regression model has been broadly utilized in the field of medicine and trauma [52,53,54] to identify the significant risk factors of the dichotomous outcome. In binary logistic regression model, the dependent variable is not limited by the assumptions of a continuous or normal distribution.

In the binary logistic regression model, the equation is formulated as follows:

$$g(x)=\beta 0+{\beta}_1{x}_1+{\beta}_2{x}_2+\dots +{\beta}_p{x}_p$$

where xj is the value of the jth independent variable, βj is the corresponding coefficient for j = 1, 2, 3,. .., p, and p is the number of independent variables.

The conditional probability of a positive outcome given the independent variable is as follows:

$$\pi (x)=\frac{\exp \left(g(x)\right)}{1+\exp \left(g(x)\right)}$$

The maximum likelihood method was used to estimate the parameters of the logistic regression model by constructing the likelihood function:

$$I\left(\beta\right)=\prod\limits_{i=1}^n\pi\left(x_i\right)^{y_i}\left(1-\mathrm\pi\left({\mathrm x}_{\mathrm i}\right)\right)^{1-y_i}$$

where yi denotes the ith observed outcome with a value of either 0 or 1 and i = 1, 2, 3,. .., n, where n is the number of observations. The best regression estimation of β was determined by maximizing the log-likelihood expression:

$$LL\left(\beta \right)=\ln \left(l\left(\beta \right)\right)=\sum\limits_{i=1}^n\left\{{y}_i\ln \left(\uppi \left({x}_i\right)\right)+\left(1-{y}_i\right)\ln \left(1-\uppi \left({x}_i\right)\right)\right\}$$

The exponentiated coefficient exp(βj), odds ratio (OR), demonstrates the effect of attributes on the likelihood of fatal injuries in logistic regression model, with a 95% confidence interval (CI) of (exp(βj − 1.96j), exp(βj + 1.96j)), where is the standard error of coefficient β. An OR of > 1 indicated a positive association between the target independent variable and fatal injuries, whereas an OR of < 1 indicated a negative association between the interest attribute and fatal injuries. An OR of 1 indicated that no association was found between the interest attributes and outcomes. If there were missing data, we conducted a sensitivity analysis to compare data with and without missing data by using the chi-square test. We used IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp to perform the statistical analysis.

Results

A total of 5,797,704 victims involved in traffic accidents were documented by police from 2011 to 2019. After applying the exclusion criteria, 2,855,996 casualties remained in the automobile–VRU crash category. Half of the casualties in automobile–VRU crashes were VRUs, and the other half were automobile drivers; accordingly, 1,427,998 VRU casualties were included in our analysis. After excluding missing data, 1,416,168 casualties with intact records were analyzed. Figure 1 illustrates the data extraction flowchart for this study. A total of 5,797,704 victims involved in traffic accidents were documented by police from 2011 to 2019. We excluded single-vehicle crashes, multiple-vehicle (> 2) crashes, VRU–VRU crashes, and crashes involving no VRUs from the dataset. Furthermore, we removed cases with missing data because we used a complete case analysis approach for our data analysis. Finally, 1,416,168 casualties with intact records were analyzed.

Table 2 presents the distribution of injury severity across a set of independent variables. Both being hit by a male driver and being a male VRU were associated with higher rates of fatal injuries to VRUs (both were 0.7%). VRUs aged ≥65 years had a higher mortality rate than other age groups. Higher than at other times, 2.9% of fatal injuries occurred at night. Regions outside a municipality (0.8%) and with a speed limit over 50 km/h (0.6%) were associated with a higher rate of fatal injuries. Nighttime with unlit streets was associated with a higher mortality rate (2.4%) than daytime. Fatal injuries were less prevalent under some road conditions, such as being a crossroad (0.6% vs. not crossroad 0.7%), a slippery road surface (0.5% vs. dry surface 0.6%), and an unobstructed view of the road (0.6% vs. with sight obstacle 1.0%). Weather did not significantly affect fatality rates. A positive BAC and a driver being unlicensed were associated with higher rates of fatal injuries among VRUs. Furthermore, the rates of fatal injuries were higher in crashes with buses and trucks (4.1%) and when the casualties were pedestrians (2.7%); bicyclists (1.6%) and motorcyclists (0.5%) had lower fatality rates. Notably, motorcyclists accounted for 91.8% of all VRUs involved in vehicle–VRU crashes. VRUs who sustained head and neck injuries had higher mortality rates (7.2%) compared with VRUs with other injured regions (0.4%).

Table 2 Distribution of injury severity among VRUs by a set of independent variables

Table 3 presents the results of the multivariate logistic models of fatal injuries. Male drivers (AOR: 1.55, 95% confidence interval [CI]: 1.45–1.67) and male VRUs (AOR: 1.63, 95% CI: 1.55–1.72) were both associated with higher risks of fatalities. VRUs aged ≥65 years were over 5 times more likely to sustain fatal injuries (AOR: 5.24, 95% CI: 4.53–6.07) than were younger groups. Fatal injuries among VRUs were more prevalent at nighttime (AOR: 4.52, 95% CI: 4.22–4.84) and in dark environments without illumination (AOR: 2.37, 95% CI: 2.05–2.75). When the travel speed was considered, counties rather than municipalities (AOR: 1.22, 95% CI: 1.16–1.28) and a speed limit ≥50 km/h (AOR: 1.29, 95% CI: 1.21–1.37) both contributed to higher likelihoods of fatal injuries. When the road surface was dry (AOR: 1.34, 95% CI: 1.24–1.44) and driver sight was obstructed (AOR: 1.39, 95% CI: 1.24–1.56), VRUs also had an additional risk of fatal injuries. Road type and road defects were not significant risk factors in multivariate analysis. Alcohol use among drivers was associated with an increased likelihood of fatal injuries to VRUs compared with alcohol nonuse. Drivers with an alcohol level ≥ 0.08% were associated with a higher VRU fatality risk (BAC 0.08–0.11%, AOR: 2.79, 95% CI: 2.14–3.63; BAC ≥ 0.11%, AOR: 2.73, 95% CI: 2.30–3.23). Unlicensed driving (AOR: 2.03, 95% CI: 1.82–2.26) and the accident involving a truck or bus (AOR: 2.82, 95% CI: 2.26–3.55) also appeared to be independent risk factors for deaths. Pedestrians (AOR: 2.17, 95% CI: 2.02–2.32) had a higher mortality rate than did other VRUs. VRUs with head and neck injuries were 12 times more likely to have fatal injuries (AOR: 12.38, 95% CI: 11.78–13.02).

Table 3 Multivariate logistic regression of VRU fatalities and potential risk factors

Table 4 presents the results of the subgroup analysis by VRU category. Drivers with a positive BAC were associated with higher odds of fatal injuries in all VRU groups. Driver BAC had a linear relationship with fatality risk among motorcyclists but not among bicyclists or pedestrians. Motorcyclists had the highest risk of death when their alcohol level was as low as 0.01–0.03% (AOR: 3.54, 95% CI: 3.08–4.08). Unlicensed riders also had a higher risk of fatalities (AOR: 1.71, 95% CI: 1.59–1.84). With regard to the effect of unlit darkness, the magnitude of the increased risk of fatal injury was the highest for pedestrians (AOR: 3.57, 95% CI: 2.68–4.76), followed by that for bicyclists (AOR: 2.66, 95% CI: 1.77–3.99) and motorcyclists (AOR: 1.55, 95% CI: 1.27–1.89).

Table 4 Subgroup analysis: fatalities of different VRUs

Discussion

Our study demonstrated that VRUs had additional risks of fatal injuries caused by drunk drivers after controlling for other variables. The higher the alcohol concentration of the driver was, the worse the fatality rates for the VRUs were, and this conclusion is in line with previous research [55,56,57,58]. A linear relationship was noted between driver BAC and the risk of fatalities among motorcyclists but not among cyclists and pedestrians. Such effects are likely attributed to several dimensions. First, in spite of speed data were not available in the Dataset, motorcyclists are generally moving much faster than those cycling or walking, thereby in turn leading to more devastating crash impacts [59,60,61,62], less reaction time [63], and high tendencies to lose control [64]. High traveling speed of motorcycles, relative to other VRUs, may act synergistically with driver BAC to increase injury severity. Such a linear relationship is likely due to the traffic exposure: fewer cyclists and pedestrians, compared with motorcycles, travel on roadways with higher speed limits. Our conjecture here needs to be ascertained in future research with additional data on crash locations and speed. While drunk driving appeared to be the main risk factors for fatal injuries among vulnerable road users, other studies [65] pointed out that mobile phone use may compromise pedestrians’ safety. Due to a lack of reliable data on mobile phone use, we identify this as a fruitful area for future studies.

Our data also highlighted that drunk riding increases motorcyclists’ mortality rate, concordant with previous research [66]. Notably, motorcyclists experienced the highest fatality rate at a legal BAC level (0.01–0.03%). In contrast to drivers, riders had the peak of fatality rate in a relative low BAC, and one early study also concluded that a low BAC level was associated with more crashes in motorcyclists than in drivers [67]. The relation between the risk of motorcyclists and their low BAC could attribute to the complexity of motorcycling, which requires concentration, balance, control and precision of movement through curves, and familiarity with the operation of the motorcycle [68]; these skills, especially balance, can be impaired at even a low alcohol concentration [69, 70]. Creaser and colleagues suggested that although riders with a low BAC preserved their cognitive and visual ability, they had to concentrate more on maintaining their riding balance, thereby sacrificing attention to cornering and hazard perception [68].

Traveling at night is generally considered risky due to poor visibility [56, 71,72,73,74,75,76]. In our data, VRUs had the highest risk of fatalities during night hours (00:00 AM to 06:59 AM), and pedestrians exhibited an additional increment in fatalities in this time frame. Compared with motorcycles and bicycles, pedestrians usually have less or no lightning instruments or reflectors, and drivers are prone to miss them in dim light. Furthermore, pedestrians also are smaller in size than other road users (i.e., machines), making them more difficult to be observed at night [77]. Appropriate measures to prevent crashes in dark environments include enhancing VRUs’ visibility through the use of lighting equipment or reflective clothes.

Head and neck injuries are common in fatal traffic crashes [78,79,80,81,82,83,84] and were associated with higher risks of death among all VRUs in our study. Helmet use reduced the fatality rate and demonstrated significant protective effects both among motorcyclists and bicyclists. The head is the only region of a rider that can be protected by a device, such as a helmet, and the benefit of a helmet in reducing injury severity and fatalities has been well documented [74, 78, 79, 81, 83]. Moreover, riding without a helmet has also been associated with other risky behaviors, such as drunk riding; both risky behaviors may lead to fatalities [85, 86]. Motorcyclists have an elevated risk of fatalities when drunk riding without helmets [52]. Although the number of head injuries in Taiwan has significantly decreased after helmet use by motorcyclists was mandated in 1997 [87], no legislation mandates the same for cyclists to date. Promoting helmet use among cyclists is clearly a public health issue.

To our knowledge, few studies have focused on the association between alcohol-impaired driving and VRU fatality. We analyzed the effect of alcohol-impaired driving on fatal injuries of VRUs and the individual risk of motorcyclists, bicyclists, and pedestrians. Our research represents to a contribution to profession through the insight that drunk driving among car drivers resulted in additional risks of mortality among all VRUs. Furthermore, a linear relationship was found between driver BAC and motorcyclist fatality rate. Interestingly, intoxicated motorcyclists, even with a BAC within the legal limit of 0.03%, had the highest rate of fatal injuries. Accordingly, we recommend several measures to improve the road safety. First, campaign for alcohol zero tolerance should be promoted to all population. Second, helmet use should not only be mandatory to motorcyclists, but also be promoted to bicyclists to reduce fatal head injuries. Furthermore, the high proportion of fatalities at nights, especially in unlit conditions, underscores the importance to enhance illumination instrument in areas where there are motorcyclists, bicyclists, and pedestrians. Last, VRUs, especially elderly bicyclists and pedestrians, may consider enhancing their own conspicuity at nights by using reflectors. However this paper is not without its limitations. First, vehicle speed was not available from the police crash records. Vehicle speed, instead of the surrogate variable “speed limit” used in the current research, may provide additional insights into fatalities. Second, detailed information on geometric factors, such as curvature or road alignments, that may play a crucial role in fatalities was not readily available from the police crash reports. Third, data on casualties who died at crash scenes were not available, and as a result, their BAC was not measured. These limitations may have underestimated the effect of drunk driving on fatalities among VRUs.

Conclusion

Drunk driving results in additional risks of mortality among all VRUs, and a linear relationship was found between driver BAC and motorcyclist fatality rate. Intoxicated motorcyclists, even with a BAC within the legal limit of 0.03%, had the highest rate of fatal injuries. The results obtained in this current research endorse a tightened legislation for alcohol concentration limit in order to prevent fatal injuries among the vulnerable road users.

Availability of data and materials

The police-reported crash data, which are open to the researchers in Taiwan, are available from the Health and Welfare Data Science Center (http://dep.mohw.gov.tw/DOS/np-2497-113.html). Only citizens of Taiwan who fulfill the requirements of conducting research projects are eligible to apply for the police-reported crash dataset. The use of police-reported crash dataset is limited to research purposes only. Applicants must follow the Computer Processed Personal Data Protection Law.

Abbreviations

BAC:

Blood-alcohol concentration

MVCs:

Motor vehicle crashes

GCS:

Glasgow Coma Scale

VRU:

Vulnerable road user

AORs:

Adjusted odds ratios

CI:

Confidence interval

References

  1. Taylor B, et al. The more you drink, the harder you fall: a systematic review and meta-analysis of how acute alcohol consumption and injury or collision risk increase together. Drug Alcohol Depend. 2010;110(1–2):108–16.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Mukamal KJ, et al. Prospective study of alcohol consumption and risk of dementia in older adults. JAMA. 2003;289(11):1405–13.

    Article  CAS  PubMed  Google Scholar 

  3. Weafer J, Fillmore MT. Acute tolerance to alcohol impairment of behavioral and cognitive mechanisms related to driving: drinking and driving on the descending limb. Psychopharmacology. 2012;220(4):697–706.

    Article  CAS  PubMed  Google Scholar 

  4. Laude JR, Fillmore MT. Simulated driving performance under alcohol: effects on driver-risk versus driver-skill. Drug Alcohol Depend. 2015;154:271–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Van Dyke N, Fillmore MT. Alcohol effects on simulated driving performance and self-perceptions of impairment in DUI offenders. Exp Clin Psychopharmacol. 2014;22(6):484.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Jordan HR, et al. Evaluating the positive drinking consequences questionnaire: support for a four-factor structure and measurement invariance. J Subst Abus. 2019;24(5):564–70.

    Google Scholar 

  7. Kloft L, et al. Hazy memories in the courtroom: a review of alcohol and other drug effects on false memory and suggestibility. Neurosci Biobehav Rev. 2021;124:291–307.

  8. Jovanovic J, et al. Vozacka sposobnost alkoholisanih vozaca motornih vozila. Facta Universitatis—Ser Med Biol. 2000;7(1):81–5.

    Google Scholar 

  9. Ristic B, et al. The influence of alcohol intoxication on the severity of injuries suffered by drivers in road traffic accidents. Eur J Trauma Emerg Surg. 2013;39(4):363–8.

    Article  CAS  PubMed  Google Scholar 

  10. Fiorentino DD. The effects of breath alcohol concentration on postural control. Traffic Injury Prev. 2018;19(4):352–7.

    Article  Google Scholar 

  11. Veldstra JL, et al. Effects of alcohol (BAC 0.5‰) and ecstasy (MDMA 100 mg) on simulated driving performance and traffic safety. Psychopharmacology. 2012;222(3):377–90.

    Article  CAS  PubMed  Google Scholar 

  12. Plawecki MH, et al. Alcohol intoxication progressively impairs drivers’ capacity to detect important environmental stimuli. Pharmacol Biochem Behav. 2018;175:62–8.

    Article  CAS  PubMed  Google Scholar 

  13. Christoforou Z, Karlaftis MG, Yannis G. Reaction times of young alcohol-impaired drivers. Accid Anal Prev. 2013;61:54–62.

    Article  PubMed  Google Scholar 

  14. George S, Rogers RD, Duka T. The acute effect of alcohol on decision making in social drinkers. Psychopharmacology. 2005;182(1):160–9.

    Article  CAS  PubMed  Google Scholar 

  15. Liu Y-C, Fu S-M. Changes in driving behavior and cognitive performance with different breath alcohol concentration levels. Traffic Inj Prev. 2007;8(2):153–61.

    Article  CAS  PubMed  Google Scholar 

  16. Du H, et al. Effects of alcohol and fatigue on driving performance in different roadway geometries. Transp Res Rec. 2016;2584(1):88–96.

    Article  Google Scholar 

  17. Price JL, et al. Effects of acute alcohol and driving complexity in older and younger adults. Psychopharmacology. 2018;235(3):887–96.

    Article  CAS  PubMed  Google Scholar 

  18. Yadav AK, Velaga NR. Effect of alcohol use on accelerating and braking behaviors of drivers. Traffic Inj Prev. 2019;20(4):353–8.

    Article  PubMed  Google Scholar 

  19. Li, J., et al., A rse-assisted gps-rss hybrid lane-level positioning system for connected vehicles. 2016.

    Google Scholar 

  20. Koval KJ, et al. The effects of alcohol on in-hospital mortality in drivers admitted after motor vehicle accidents. Bull NYU Hosp Jt Dis. 2008;66(1):27–34.

    PubMed  Google Scholar 

  21. Chang S, Cushman JG, Pasquale MD. The injured intoxicated driver: analysis of the conviction process. J Trauma Acute Care Surg. 2001;51(3):551–6.

    Article  CAS  Google Scholar 

  22. Castano R. The drink driving situation in Colombia. Traffic Inj Prev. 2012;13(2):120–5.

    Article  PubMed  Google Scholar 

  23. Ogazi C, Edison E. The drink driving situation in Nigeria. Traffic Inj Prev. 2012;13(2):115–9.

    Article  PubMed  Google Scholar 

  24. Ngoc LB, Thieng NT, Huong NL. The drink driving situation in Vietnam. Traffic Inj Prev. 2012;13(2):109–14.

    Article  PubMed  Google Scholar 

  25. Li Y, et al. The drink driving situation in China. Traffic Inj Prev. 2012;13(2):101–8.

    Article  PubMed  Google Scholar 

  26. Mukamal, K.J., Overview of the risks and benefits of alcohol consumption. UpToDate, Basow, DS (Ed), UpToDate, Waltham, 2010.

  27. C Verster J, et al. The alcohol hangover research group consensus statement on best practice in alcohol hangover research. Curr Drug Abuse Rev. 2010;3(2):116–26.

    Article  Google Scholar 

  28. Hadjizacharia P, et al. Alcohol exposure and outcomes in trauma patients. Eur J Trauma Emerg Surg. 2011;37(2):169–75.

    Article  CAS  PubMed  Google Scholar 

  29. Ahmed N, Greenberg P. Patient mortality following alcohol use and trauma: a propensity-matched analysis. Eur J Trauma Emerg Surg. 2019;45(1):151–8.

    Article  CAS  PubMed  Google Scholar 

  30. Pasnin LT, Gjerde H. Alcohol and drug use among road users involved in fatal crashes in Norway. Traffic Inj Prev. 2021;22(4):267–71.

  31. Moskowitz H, et al. Driver characteristics and impairment at various BACs. United States: National Highway Traffic Safety Administration; 2000.

    Book  Google Scholar 

  32. Moskowitz H, Florentino D. A review of the literature on the effects of low doses of alcohol on driving-related skills. United States: National Highway Traffic Safety Administration; 2000.

    Book  Google Scholar 

  33. Fell JC, Scherer M. Estimation of the potential effectiveness of lowering the blood alcohol concentration (BAC) limit for driving from 0.08 to 0.05 grams per deciliter in the United States. Alcohol Clin Exp Res. 2017;41(12):2128–39.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Martin J-L, et al. Cannabis, alcohol and fatal road accidents. Plos One. 2017;12(11):e0187320.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Phillips DP, Brewer KM. The relationship between serious injury and blood alcohol concentration (BAC) in fatal motor vehicle accidents: BAC= 0.01% is associated with significantly more dangerous accidents than BAC= 0.00%. Addiction. 2011;106(9):1614–22.

    Article  PubMed  Google Scholar 

  36. Jinghong L, Yongtengye S. Effects of a small self-control alcohol dose on skilled drivingperformance. Psychol Sci. 1999;22:120–2.

    Google Scholar 

  37. Zhang X, et al. Effect of different breath alcohol concentrations on driving performance in horizontal curves. Accid Anal Prev. 2014;72:401–10.

    Article  PubMed  Google Scholar 

  38. Culhane J, Silverglate B, Freeman C. Alcohol is a predictor of mortality in motor vehicle collisions. J Saf Res. 2019;71:201–5.

    Article  Google Scholar 

  39. Sutlovic D, et al. The role of alcohol in road traffic accidents with fatal outcome: 10-year period in Croatia Split–Dalmatia County. Traffic Inj Prev. 2014;15(3):222–7.

    Article  PubMed  Google Scholar 

  40. Zador PL, Krawchuk SA, Voas RB. Alcohol-related relative risk of driver fatalities and driver involvement in fatal crashes in relation to driver age and gender: an update using 1996 data. J Stud Alcohol. 2000;61(3):387–95.

    Article  CAS  PubMed  Google Scholar 

  41. Reynaud M, et al. Alcohol is the main factor in excess traffic accident fatalities in France. Alcohol Clin Exp Res. 2002;26(12):1833–9.

    Article  PubMed  Google Scholar 

  42. Philip P, et al. Fatigue, alcohol, and serious road crashes in France: factorial study of national data. BMJ. 2001;322(7290):829–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Siliquini R, et al. A European study on alcohol and drug use among young drivers: the TEND by night study design and methodology. BMC Public Health. 2010;10(1):1–6.

    Article  Google Scholar 

  44. Barry V, Schumacher A, Sauber-Schatz E. Alcohol-impaired driving among adults—USA, 2014–2018. Inj Prev. 2022;28(3):211–7.

    Article  PubMed  Google Scholar 

  45. Høye A. Speeding and impaired driving in fatal crashes—results from in-depth investigations. Traffic Inj Prev. 2020;21(7):425–30.

    Article  PubMed  Google Scholar 

  46. Yadav AK, Velaga NR. Alcohol-impaired driving in rural and urban road environments: effect on speeding behaviour and crash probabilities. Accid Anal Prev. 2020;140:105512.

    Article  PubMed  Google Scholar 

  47. Rifaat SM, Tay R, De Barros A. Effect of street pattern on the severity of crashes involving vulnerable road users. Accid Anal Prev. 2011;43(1):276–83.

    Article  PubMed  Google Scholar 

  48. Cercarelli LR, et al. Travel exposure and choice of comparison crashes for examining motorcycle conspicuity by analysis of crash data. Accid Anal Prev. 1992;24(4):363–8.

    Article  CAS  PubMed  Google Scholar 

  49. Wulf G, Hancock P, Rahimi M. Motorcycle conspicuity: an evaluation and synthesis of influential factors. J Saf Res. 1989;20(4):153–76.

    Article  Google Scholar 

  50. Rogé J, et al. Mechanisms underlying cognitive conspicuity in the detection of cyclists by car drivers. Accid Anal Prev. 2017;104:88–95.

    Article  PubMed  Google Scholar 

  51. Wood JM. Nighttime driving: visual, lighting and visibility challenges. Ophthalmic Physiol Opt. 2020;40(2):187–201.

    Article  PubMed  Google Scholar 

  52. Wiratama BS, et al. Evaluating the combined effect of alcohol-involved and un-helmeted riding on motorcyclist fatalities in Taiwan. Accid Anal Prev. 2020;143:105594.

    Article  PubMed  Google Scholar 

  53. Chen P-L, Pai C-W. Evaluation of injuries sustained by motorcyclists in approach-turn crashes in Taiwan. Accid Anal Prev. 2019;124:33–9.

    Article  CAS  PubMed  Google Scholar 

  54. Chan T-C, et al. Association of air pollution and weather factors with traffic injury severity: a study in Taiwan. Int J Environ Res Public Health. 2022;19(12):7442.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Mohamed MG, et al. A clustering regression approach: a comprehensive injury severity analysis of pedestrian–vehicle crashes in New York, US and Montreal, Canada. Saf Sci. 2013;54:27–37.

    Article  Google Scholar 

  56. Pai C-W, et al. Walking against or with traffic? Evaluating pedestrian fatalities and head injuries in Taiwan. BMC Public Health. 2019;19(1):1–11.

    Article  Google Scholar 

  57. Moudon AV, et al. The risk of pedestrian injury and fatality in collisions with motor vehicles, a social ecological study of state routes and city streets in King County, Washington. Accident Analysis Prev. 2011;43(1):11–24.

    Article  Google Scholar 

  58. Kim J-K, et al. A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model. Accid Anal Prev. 2010;42(6):1751–8.

    Article  PubMed  Google Scholar 

  59. Elvik R, Christensen P, Amundsen AH. Speed and road accidents: an evaluation of the Power Model: Transportøkonomisk Institutt; 2004.

    Google Scholar 

  60. Komol MMR, et al. Crash severity analysis of vulnerable road users using machine learning. Plos One. 2021;16(8):e0255828.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Batomen B, et al. Vulnerable road-user deaths in Brazil: a Bayesian hierarchical model for spatial-temporal analysis. Int J Inj Control Saf Promot. 2020;27(4):528–36.

    Article  Google Scholar 

  62. Verma V, et al. Epidemiology of trauma victims admitted to a level 2 trauma center of North India. Int J Crit Illn Inj Sci. 2017;7(2):107.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Murphy P, Morris A. Quantifying accident risk and severity due to speed from the reaction point to the critical conflict in fatal motorcycle accidents. Accid Anal Prev. 2020;141:105548.

    Article  PubMed  Google Scholar 

  64. Van Elslande P, Elvik R. Powered two-wheelers within the traffic system: Elsevier; 2012;49:1–4.

  65. Useche SA, Alonso F, Montoro L. Validation of the walking behavior questionnaire (WBQ): a tool for measuring risky and safe walking under a behavioral perspective. J Transp Health. 2020;18:100899.

    Article  Google Scholar 

  66. Ahmed N, et al. Elevated blood alcohol impacts hospital mortality following motorcycle injury: a National Trauma Data Bank analysis. Injury. 2020;51(1):91–6.

    Article  PubMed  Google Scholar 

  67. Sun SW, Kahn DM, Swan KG. Lowering the legal blood alcohol level for motorcyclists. Accid Anal Prev. 1998;30(1):133–6.

    Article  CAS  PubMed  Google Scholar 

  68. Creaser JI, et al. Effects of alcohol impairment on motorcycle riding skills. Accid Anal Prev. 2009;41(5):906–13.

    Article  PubMed  Google Scholar 

  69. Rudin-Brown CM, et al. Performance of a cognitive, but not visual, secondary task interacts with alcohol-induced balance impairment in novice and experienced motorcycle riders. Accid Anal Prev. 2013;50:895–904.

    Article  PubMed  Google Scholar 

  70. Franks H, et al. The relationship between alcohol dosage and performance decrement in humans. J Stud Alcohol. 1976;37(3):284–97.

    Article  CAS  PubMed  Google Scholar 

  71. Pai C-W. Motorcycle right-of-way accidents—a literature review. Accid Anal Prev. 2011;43(3):971–82.

    Article  PubMed  Google Scholar 

  72. Wood JM, et al. Bicyclists overestimate their own night-time conspicuity and underestimate the benefits of retroreflective markers on the moveable joints. Accid Anal Prev. 2013;55:48–53.

    Article  PubMed  Google Scholar 

  73. Zahabi SAH, et al. Estimating potential effect of speed limits, built environment, and other factors on severity of pedestrian and cyclist injuries in crashes. Transp Res Rec. 2011;2247(1):81–90.

    Article  Google Scholar 

  74. Rifaat SM, Tay R, De Barros A. Severity of motorcycle crashes in Calgary. Accid Anal Prev. 2012;49:44–9.

    Article  PubMed  Google Scholar 

  75. Pai C-W. Motorcyclist injury severity in angle crashes at T-junctions: identifying significant factors and analysing what made motorists fail to yield to motorcycles. Saf Sci. 2009;47(8):1097–106.

    Article  Google Scholar 

  76. Quddus MA, Noland RB, Chin HC. An analysis of motorcycle injury and vehicle damage severity using ordered probit models. J Saf Res. 2002;33(4):445–62.

    Article  Google Scholar 

  77. Kwan I, Mapstone J. Interventions for increasing pedestrian and cyclist visibility for the prevention of death and injuries. Cochrane Database Syst Rev. 2006;4. Art. No.: CD003438.

  78. Leijdesdorff HA, et al. Injury pattern, injury severity, and mortality in 33,495 hospital-admitted victims of motorized two-wheeled vehicle crashes in the Netherlands. J Trauma Acute Care Surg. 2012;72(5):1363–8.

    Article  PubMed  Google Scholar 

  79. Solagberu B, et al. Motorcycle injuries in a developing country and the vulnerability of riders, passengers, and pedestrians. Inj Prev. 2006;12(4):266–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  80. Lip HTC, et al. Clinical characteristics of 1653 injured motorcyclists and factors that predict mortality from motorcycle crashes in Malaysia. Chin J Traumatol. 2019;22(2):69–74.

    Article  Google Scholar 

  81. Pai C-W, et al. A population-based case-control study of hospitalisation due to head injuries among bicyclists and motorcyclists in Taiwan. BMJ Open. 2017;7(11):e018574.

    Article  PubMed  PubMed Central  Google Scholar 

  82. Depreitere B, et al. Bicycle-related head injury: a study of 86 cases. Accid Anal Prev. 2004;36(4):561–7.

    Article  PubMed  Google Scholar 

  83. Mayrose J. The effects of a mandatory motorcycle helmet law on helmet use and injury patterns among motorcyclist fatalities. J Saf Res. 2008;39(4):429–32.

    Article  Google Scholar 

  84. Juhra C, et al. Bicycle accidents–do we only see the tip of the iceberg?: a prospective multi-Centre study in a large German city combining medical and police data. Injury. 2012;43(12):2026–34.

    Article  CAS  PubMed  Google Scholar 

  85. Dos Santos WJ, et al. Alcohol and risky behavior in traffic among motorcyclists involved in accidents in a city in northeastern Brazil. Traffic Inj Prevn. 2019;20(3):233–7.

    Article  Google Scholar 

  86. Tongklao A, Jaruratanasirikul S, Sriplung H. Risky behaviors and helmet use among young adolescent motorcyclists in southern Thailand. Traffic Inj Prev. 2016;17(1):80–5.

    Article  PubMed  Google Scholar 

  87. Chiu W-T, et al. Implementation of a motorcycle helmet law in Taiwan and traffic deaths over 18 years. JAMA. 2011;306(3):267–8.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

This manuscript was edited by Wallace Academic Editing.

Funding

This study was financially supported by grants from the Ministry of Science and Technology, Taiwan (MOST 110–2410-H-038-016-MY2 and MOST 109–2314-B-038-066-) and New Taipei City Hospital (NTPC111–004). The funders had no role in the design of the study, data collection and analysis, interpretation of data, or preparation of the manuscript.

Author information

Authors and Affiliations

Authors

Contributions

HAL was in charge of data analysis, interpretation, reporting, and original draft preparation. CWC reviewed the relevant literature, interpreted and analysed the data, and edited the manuscript. BSW was in charge of data extraction and analysis. PLC provided study conceptualization and supervision. MHW and CJC reviewed the literature and revised the manuscript. WS supported the literature search and quality assessment. HCH reviewed the literature and edited the manuscript. CWP was responsible for study design, data interpretation, drafted the manuscript, and supervision. The final version of the manuscript was read and approved by all contributing authors.

Corresponding author

Correspondence to Chih-Wei Pai.

Ethics declarations

Ethics approval and consent to participate

The current research analysed national crash data without individuals’ confidential information such as names or identity numbers. Consent to participate was therefore not requested. All methods were carried out in accordance with relevant guidelines and regulations. The study was approved in its entirety by the Institutional Review Board affiliated with Taipei Medical University (IRB#: N202007045). Administrative permission was required to access the raw data from the “National Taiwan Traffic Crash Dataset”. The Health and Welfare Data Science Center grants permission.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lin, HA., Chan, CW., Wiratama, B.S. et al. Evaluating the effect of drunk driving on fatal injuries among vulnerable road users in Taiwan: a population-based study. BMC Public Health 22, 2059 (2022). https://doi.org/10.1186/s12889-022-14402-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12889-022-14402-3

Keywords