Traffic Inj Prev [JOURNAL]
Sun
200 papers
Evaluation of the redesigned "Booze It & Lose It" campaign in North Carolina.
Farmer CM, Eichelberger AH, Ezzell MM … +1 more , Gianni TJ
Traffic Inj Prev · 2026 Jan · PMID 41556774 · Publisher ↗
OBJECTIVE: In 2023, the North Carolina Governor's Highway Safety Program and the Insurance Institute for Highway Safety joined with local law enforcement and other partners to update the "Booze It & Lose It" campaign thr...
OBJECTIVE: In 2023, the North Carolina Governor's Highway Safety Program and the Insurance Institute for Highway Safety joined with local law enforcement and other partners to update the "Booze It & Lose It" campaign through a pilot project in western North Carolina. This article evaluates the effects of the revised Booze It & Lose It campaign on driver awareness and behavior and the numbers of serious roadway crashes. METHODS: Drivers were surveyed in both the pilot test communities and comparison communities in North Carolina to assess campaign awareness, beliefs, and self-reported behaviors. Police-reported crashes for 2015-2024 were used to examine the impact of the program on the rates of overall crashes, single-vehicle nighttime crashes, single-vehicle late-night crashes, and serious and fatal injury crashes using time-series regressions. Crash data were adjusted for seasonality and the possible effects of COVID-19. RESULTS: The percentage of drivers in the pilot counties who were very or somewhat familiar with Booze It & Lose It increased from 66% before the campaign to 68% after 1 year and 81% after 2 years, while the percentage in the comparison counties fluctuated from 76% to 72% after 1 year and 82% after 2 years. The estimated difference in trends across county groups is not significantly different ( = .0653). The pilot program was associated with a relative 4.3% increase in all crashes, a 1.1% decrease in single-vehicle nighttime crashes, a 6.9% decrease in single-vehicle late night crashes, and a 14.4% decrease in serious and fatal injury crashes, but none of these estimates are statistically significant. CONCLUSIONS: Driving while impaired by alcohol remains an issue in North Carolina. If the revised Booze It & Lose It pilot program changed driver behavior and subsequent alcohol-related crash rates during these 2 years, the change was too small to confidently detect. There are, however, some encouraging signs. Awareness of the Booze It & Lose It program has rebounded to a level not seen since 2012. And crash types typically associated with alcohol are down. Maybe most importantly, the pilot program is continuing under local control.
Speed limits and fatality risk in Australia.
Doecke SD, Baldock MRJ
Traffic Inj Prev · 2026 Jan · PMID 41556769 · Publisher ↗
OBJECTIVE: To assess the degree to which speed limits are being set in accordance with the Safe System, with fatal crash rates remaining negligible regardless of speed limit. A further objective was to examine the reduct...
OBJECTIVE: To assess the degree to which speed limits are being set in accordance with the Safe System, with fatal crash rates remaining negligible regardless of speed limit. A further objective was to examine the reduction in fatal crashes that could be achieved if the fatal crash rate were equal to the lowest fatality rate in a common speed zone. METHOD: Police reported crashes from 2016 to 2023 in two Australian states were used to calculate the fatal crash rate by speed limit. The analysis was conducted at three levels; combined data, by state, and by state and area. Fatal crash reductions were projected based on all speed zones having a fatal crash rate equal to that for roads with a speed limit of 50 km/h. RESULTS: The fatal crash rates were found to be non-negligible at all speed limits. Furthermore, the fatal crash rates generally increased as speed limits increased. This suggests that the speed limits in Australia do not sufficiently account for the different levels of risk found on the road network. In an 80 km/h zone, the risk of a crash being fatal is three times higher than in a 50 km/h zone and, in a 100 km/h zone, it is more than six times that of a 50 km/h zone. Of particular concern is the 110 km/h zones in South Australia, where the risk of a crash being fatal is more than 12 times higher than in a 50 km/h zone. It was found that, if the fatal crash rate for all speed zones was equal to the rate in 50 km/h zones, the number of fatal crashes would be reduced by between 46.2% and 55.5%. CONCLUSIONS: Speed limits in Australia are not sufficiently matched to the risk to road users should a crash occur, with substantial increases in the fatal crash rate at speeds limits of 80 km/h and above. A uniform fatal crash rate equal to the fatal crash rate found in 50 km/h zones would halve the number of fatal crashes. Lowering the fatal crash rate in Australia could be achieved through improvements to the roads, improvements in the passive safety of vehicles, and the lowering of speed limits, or a combination of these approaches.
A machine learning model to predict the severity of road traffic injury based on aberrant driving behaviors and driver characteristics.
Abdolrazaghi A, Zarabadi Pour S
Traffic Inj Prev · 2026 Jan · PMID 41556763 · Publisher ↗
OBJECTIVES: Road traffic injuries remain a leading cause of mortality and disability worldwide, especially in low- and middle-income countries. This study aimed to develop and validate a machine-learning model to predict...
OBJECTIVES: Road traffic injuries remain a leading cause of mortality and disability worldwide, especially in low- and middle-income countries. This study aimed to develop and validate a machine-learning model to predict road traffic injury severity, using aberrant driving behaviors as measured by the Manchester Driver Behavior Questionnaire, along with demographic and driving exposure variables. METHODS: We conducted a secondary analysis of a dataset including 800 drivers. Participants were categorized into three road traffic injury severity classes: no injury ( = 400), mild injury ( = 200), and severe or fatal injury ( = 200). The Boruta feature selection algorithm was applied to confirm predictor relevance. An XGBoost classification model was trained and tuned through 10-fold cross-validation. Model performance was evaluated using multiclass AUC, accuracy, F1 score, and class-specific diagnostic indices. RESULTS: The optimized XGBoost model achieved a multiclass AUC of 0.886 and an overall accuracy of 74.4%. Sensitivity was highest for predicting no injury (0.883), while specificity peaked for the severe/fatal injury class (0.910). The most influential predictors included crash history, education level, and daily driving hours. Among behavioral variables, slips and deliberate violations emerged as meaningful contributors to injury severity predictions. CONCLUSIONS: Integrating psychological constructs from the Driver Behavior Questionnaire with demographic and exposure data provides predictive accuracy in injury severity models. Our behavior-informed framework offers a practical and theoretically grounded tool for identifying high-risk drivers and informing targeted road safety interventions.
Identifying the contributory chains and patterns of road facilities in bus-involved crashes using latent class clustering and association rules.
Nie C, Wang S, Lin Q
Traffic Inj Prev · 2026 Jan · PMID 41556753 · Publisher ↗
OBJECTIVE: This study aims to investigate the heterogeneity in bus-involved crashes by identifying association rules and contributory patterns of road facilities across different crash types and severities. The goal is t...
OBJECTIVE: This study aims to investigate the heterogeneity in bus-involved crashes by identifying association rules and contributory patterns of road facilities across different crash types and severities. The goal is to support more targeted and context-specific safety interventions. METHODS: A total of 14,560 bus-involved crashes that occurred in London, UK, from 2010 to 2019 were analyzed. Each crash record includes attributes related to the driver, vehicle, environment, and road infrastructure. To address crash heterogeneity, Latent Class Clustering (LCC) was used to classify crashes into distinct clusters based on uncontrollable factors such as driver characteristics and environmental conditions. Subsequently, Association Rule Mining (ARM) was applied to identify contributory chains and association patterns of controllable road infrastructure factors within each cluster and severity level. The top ten rules for both slight and serious crashes were extracted in each group. RESULTS: The findings reveal that typical features and association rules vary significantly across crash clusters and severity levels. On one hand, a single associative chain can result in varying severity levels across different crash clusters. For example, dual carriageways and daylight are contributing factors to both slight and severe crashes. On the other hand, associative chains leading to the same severity level may differ among crash clusters. For instance, yield or uncontrolled junctions, zebra crossings, and daylight are key factors in slight crashes within Cluster 3, but not in other clusters. CONCLUSIONS: The proposed framework, which integrates LCC and ARM, enables a more nuanced analysis of bus-involved crash data by separating classification and association rules analysis based on controllability. The results provide actionable insights into how specific road infrastructure features contribute to different types and severities of crashes. This can inform the design of targeted safety measures and infrastructure improvements to reduce crash risks in urban bus systems. Moreover, the framework is adaptable to other transportation modes and crash datasets.
A finite element study on the effect of initial head-neck posture on neck injury risk in frontal collisions.
Liu Z, Zhang XY, Liu W … +3 more , Wu Y, Du CF, Zhang Q
Traffic Inj Prev · 2026 Jan · PMID 41556749 · Publisher ↗
OBJECTIVE: Neck injuries represent a predominant category of injuries in automotive collisions. Non-neutral head-neck postures are believed to increase the risk of neck injury during rear impact scenarios. However, quant...
OBJECTIVE: Neck injuries represent a predominant category of injuries in automotive collisions. Non-neutral head-neck postures are believed to increase the risk of neck injury during rear impact scenarios. However, quantitative research on how initial head-neck postures of front-seat passengers influence neck injury mechanisms in frontal collisions remains scarce. Hence, the present study aimed to investigate the effects of different head-neck postures on neck injury risk for front-seat passengers during frontal collisions. METHODS: A finite element model of the occupant restraint system was developed using the Total Human Model for Safety (THUMS) 50th percentile male dummy. Under frontal collision conditions (impact velocity: 50 km/h; acceleration pulse duration: 0.12 s), eight head-neck postures were evaluated: neutral (0°), axial rotation (15°, 30°, 45°, 60°), flexion (15°), extension (15°), and lateral bending (15°). RESULTS: Compared with the neutral posture, non-neutral head-neck postures significantly increased vertebral stress, intervertebral disc pressure, ligament strain, NIC, N and N. Among the seven neck ligaments analyzed, the capsular ligaments consistently exhibited the highest strain across all postures. CONCLUSIONS: Non-neutral head-neck postures during frontal collisions increase cervical biomechanical loading compared with the neutral posture. NIC and Nkm exceeded their injury thresholds in all non-neutral postures, while Nij remained below its threshold but exhibited a similar trend. Among the postures studied, extension and high axial rotation resulted in the highest loading on cervical soft tissues. Maintaining a neutral posture may help reduce cervical injury risk for front-seat passengers.
Community norms and transportation safety behaviors among caregivers of children 6 months to 10 years old in chicago.
Landa MM, Pollock B, Castillo L … +3 more , Guillaume A, Ibrahim S, Macy ML
Traffic Inj Prev · 2026 Jan · PMID 41537774 · Full text
OBJECTIVE: Motor vehicle collisions are a leading cause of death for children. Risks of driver speeding, distraction, and impairment and protective benefits of following child passenger safety guidelines are well establi...
OBJECTIVE: Motor vehicle collisions are a leading cause of death for children. Risks of driver speeding, distraction, and impairment and protective benefits of following child passenger safety guidelines are well established, as are differences in safety behaviors and crash outcomes for child passengers across racial, ethnic, and linguistic groups. Persistent disparities warrant contemporary exploration of potentially modifiable factors across groups. In this study we examine differences in community transportation safety norms in relation to caregiver-reported risky driving behaviors, restraint use, and child passenger safety guideline adherence across racial, ethnic, and linguistic groups. METHODS: We conducted a secondary analysis of baseline data collected in the Tiny Cargo, Big Deal/Abróchame Bien, Cuídame Bien clinical trial (NCT04238247). We examined caregiver-reported agreement with eight statements about community norms for transportation safety (e.g., acceptability of cell phone use and adherence to child passenger safety guidelines) and their self-reported engagement in six risky driving behaviors (e.g., speeding and cell phone distractions), two seat belt use situations, and three child passenger safety behaviors. In complete case chi-square and Fisher's exact tests analyses, we compared transportation behaviors and community norms across race, ethnicity, and preferred language. We also tested for associations between agreement with community norm statements and caregiver-reported engagement in related transportation safety behaviors using unadjusted logistic regression and logistic regression adjusted for caregiver demographic characteristics and child age. Analyses were conducted in Stata 18.0. RESULTS: Complete responses from 474 caregivers of children 6-months to 10-years-old were analyzed. Most participants identified as mothers (93.2%), 64.1% were married, and 60.97% completed a bachelor's, graduate or professional degree. The median caregiver age was 36 years (interquartile range (IQR) 21, 50). One in ten participants preferred Spanish ( = 52) and among caregivers who preferred English, most identified as white ( = 175) or Hispanic/Latine ( = 138). Racial, ethnic, and linguistic group variation was observed for ten of eleven transportation behaviors and six of eight community norms for transportation safety ( < 0.05). Traveling unrestrained was most common among caregivers who preferred Spanish. However, speeding and cell phone use while driving were more common among caregivers who identified as white and preferred English. Several risky transportation behaviors were associated with related community norms (e.g., caregivers who perceived texting while driving as unacceptable in their community had higher odds of reporting never checking their phone while driving and their car was moving) (Adjusted Odds Ratio 3.61, 95% Confidence Interval: 2.18, 5.99, < 0.001). CONCLUSIONS: Caregiver reported engagement in risky transportation behaviors varied by race, ethnicity, and preferred language and was associated with community norms. Linguistically and culturally tailored campaigns seeking to modify community norms may be a strategy to reduce risky transportation behaviors among caregivers of young children.
Driving under the influence of drugs: perceptions and attitudes of Chinese professional drivers.
Wang J, Jia S, Li J … +4 more , Huang M, Lu B, Pei X, Shen S
Traffic Inj Prev · 2026 Jan · PMID 41537766 · Publisher ↗
OBJECTIVES: This study explored the status and causative factors of drug driving behavior among professional drivers in China, by examining their perceptions of impairment caused by legal and illegal drugs, attitudes tow...
OBJECTIVES: This study explored the status and causative factors of drug driving behavior among professional drivers in China, by examining their perceptions of impairment caused by legal and illegal drugs, attitudes toward police enforcement and legal regulation, and factors predicting the behavior. METHODS: The current study recruited a sample of professional Liquefied Natural Gas (LNG) transport drivers completing a designed questionnaire. A total of 294 individuals, ranging in age between 19 and 58 years ( = 48.03, = 4.46) are engaged. Data were analyzed using ANOVA, Path Analysis, and Bayesian Network Learning. RESULTS: The participants rated illegal drugs as more dangerous than legal ones, of which the full list is provided in the online supplement. One-way ANOVA revealed significant differences in impairment ratings across legal drug categories ( < .001), with alcohol (D1) perceived as the most impairing. Path Analysis identified that the perceived risk of drug driving (β = 0.26, < .001) and attitude toward legal supervision (β = 0.23, < .001) were significant direct predictors of self-reported drug driving behavior. Subsequently, the Bayesian Network model quantified this relationship, showing that drivers who perceived the highest level of risk had a substantially lower probability (31.6%) of engaging in drug driving. Two-thirds of respondents supported enhanced legal supervision for this issue. CONCLUSIONS: The findings highlight distinct perceptions and attitudes in this population, suggesting that targeted education and legal measures could effectively manage drug driving in the Chinese context.
Research on occupants out-of-position and whiplash injury under automatic emergency braking.
Huang J, Jiang C, Wang G … +4 more , Shi L, Zhang G, Shi A, Liang R
Traffic Inj Prev · 2026 Jan · PMID 41534037 · Publisher ↗
OBJECTIVES: Automobile manufacturers are gradually incorporating autonomous emergency braking (AEB) systems into their vehicles. However, AEB may result in the occupant being out of position, increasing the risk of injur...
OBJECTIVES: Automobile manufacturers are gradually incorporating autonomous emergency braking (AEB) systems into their vehicles. However, AEB may result in the occupant being out of position, increasing the risk of injury. Additionally, the accident analysis of occupant injury in the scenario where AEB Works First, Followed by Rear Impact (AWFFRI), is limited. This article aims to investigate the occupants' out-of-position response and whiplash injury in AWFFRI. METHODS: The Total Human Model for Safety (THUMS), which was proposed and validated by the Toyota company, was used to provide a baseline for human injury. Furthermore, THUMS and BioRID II dummy models were utilized to develop numerical sled test models. The BioRID II sled test finite element (FE) models were developed and refined through a physical experiment. Then, the difference between BioRID II and THUMS dummy kinematics in the out-of-position scenario was analyzed. The force, tension, acceleration, NIC, strain and stress of the BioRID II and THUMS models in traditional and AWFFRI situations were measured. The effect of AEB on each dummy was analyzed. RESULTS: The BioRID II dummy exhibits a 19 mm lower head displacement than the THUMS model, resulting in a 13% difference. In the AWFFRI scenario, the AEB, which results in a -8.8 m/s lowest acceleration, will last for 0.342 s, followed by a rear-end collision with 97.6 m/s maximum acceleration. The BioRID II and THUMS models in the AWFFRI situation illustrated higher whiplash injuries than the traditional scenario, with the upper neck shear force (164.79 N), upper neck extension tension (830.46 N), T1 average acceleration (137.84 m/s), NIC (45.68), and interspinous ligament strain (0.871) increased by 158.45%, 96.62%, 7.23%, 89.70% and 108.9%. CONCLUSIONS: The BioRID II dummy response in the AWFFRI scenario exhibit an increasing trend compared to those in the traditional whiplash scenario. For the BioRID II dummy, the upper neck shear force, upper neck extension tension, as well as NIC exhibit the most significant increase. Additionally, the upper neck shear force, upper neck extension tension, T1 average acceleration, and NIC of the BioRID II dummy exceeded the thresholds, indicating that the occupant may be at risk of sustaining a neck injury. For the THUMS model, the stress and strain on the dummy skull, cervical vertebrae, and ligaments are higher in the AWFFRI scenario compared to the traditional whiplash scenario. Furthermore, the interspinous ligament injury stain (0.871) exceeds the threshold (0.68), indicating a greater risk of ligamentum flavum injury. The results obtained in this study highlight a concern for a potential increase in whiplash risk associated with AEB when followed by a rear impact.
Optimization and validation of multiscale feature selection for EEG-based recognition of drivers' negative emotions.
Li Y, Tang B, Luo B … +2 more , Wu Y, Ding Y
Traffic Inj Prev · 2026 Jan · PMID 41532989 · Publisher ↗
OBJECTIVE: Negative emotions, such as stress and anger, are significant factors leading to dangerous driving behavior. Investigating the impact of these emotions on driving safety is crucial for effective traffic injury...
OBJECTIVE: Negative emotions, such as stress and anger, are significant factors leading to dangerous driving behavior. Investigating the impact of these emotions on driving safety is crucial for effective traffic injury prevention. METHODS: Electroencephalography (EEG) is a valuable tool for detecting emotional neural activity due to its high temporal resolution and noninvasive characteristics. This study systematically analyzes the multiscale characteristics of EEG signals across different frequency bands, brain regions, and feature domains to elucidate the neural mechanisms of stress and anger states during driving. EEG signals were collected from 32 participants in a driving simulation environment under both emotional (stress and anger) and nonemotional conditions. First, time domain features (mean, variance, skewness, kurtosis), frequency domain features (power spectral density, Shannon entropy), and spatial domain features (common spatial patterns) were extracted from five key brain regions-frontal pole, frontal, central, parietal, and occipital-and across five frequency bands ranging from delta to gamma. Second, feature selection was performed through correlation analysis, and a support vector machine classifier was utilized for machine learning validation. RESULTS: The results indicate that the frontal pole region exhibited the highest frequency of neural activity, while frequency band analysis revealed that beta wave activity occurred most frequently. Furthermore, common spatial patterns demonstrated a significant ability to differentiate between stress and anger emotions. The optimal mixed feature set achieved classification accuracies of 85% for stress and 80% for anger, representing improvements of 20% to 55% and 7% to 55%, respectively, over the unselected feature set, as well as enhancements of 10% to 25% and 10% to 30%, respectively, compared to single-domain features. CONCLUSIONS: These findings provide a neurophysiological basis for developing a wearable EEG-based driver emotion recognition system that has the potential to enhance traffic safety and reduce the risk of accidents by enabling timely interventions based on drivers' emotional states.
Safety impact study of 3D linear guidance signs on desert highways.
Jia X, Fu Y, Liu S … +3 more , Qu Y, Yao H, Wang A
Traffic Inj Prev · 2026 Jan · PMID 41505770 · Publisher ↗
OBJECTIVES: In view of the traffic safety problems caused by insufficient line-of-sight guidance in special linear sections of desert highways, this study carries out the optimization design research of 3D linear guidanc...
OBJECTIVES: In view of the traffic safety problems caused by insufficient line-of-sight guidance in special linear sections of desert highways, this study carries out the optimization design research of 3D linear guidance signs based on the visual perception theory in order to improve the effect of visual guidance in desert environments. METHODS: According to the characteristics of the desert environment, two conventional linear guide signs are used as the control group, and three 3D linear guidance signs are used as the experimental group. Through the combination of driving simulation experiments and real-vehicle tests, the systematically collected data of drivers' speed regulation, lateral offset, pupil diameter change, fixation time and subjective evaluation indexes. The index data were compared and analyzed, and a comprehensive evaluation model was constructed to select the optimal line mark. RESULTS: In terms of visual guidance efficacy, Option 1 reduced the average speed by 6.05 km/h compared to Control 2 and decreased the standard deviation of lateral offset to 0.267 m. It also resulted in a more stable average pupil diameter of 3.28 mm and a maximum fixation duration of 473.74 ms. Multi-dimensional evaluation showed that Option 1 achieved the best overall performance, with all metrics significantly outperforming other options. CONCLUSIONS: The study shows that the three-dimensional stereoscopic design effectively improves the visual guidance effect under the desert environment by enhancing the spatial hierarchy and directional indication of the sign, and the research results provide the theoretical basis and technical support for the optimization of traffic safety facilities on desert highways.
Belted driver fatalities in oblique frontal crashes.
Viano DC
Traffic Inj Prev · 2026 Jan · PMID 41505767 · Publisher ↗
OBJECTIVE: This study analyzed fatal crashes involving lap-shoulder belted drivers in oblique frontal crashes. The front crush and occupant compartment damage of newer vehicles were compared with NHTSA and IIHS crash tes...
OBJECTIVE: This study analyzed fatal crashes involving lap-shoulder belted drivers in oblique frontal crashes. The front crush and occupant compartment damage of newer vehicles were compared with NHTSA and IIHS crash tests. METHODS: NHTSA online database of 2017-2023 CISS (Crash Investigation Sampling System) crashes was searched for driver fatalities in oblique frontal crashes with ±10° or ±20° PDOF and delta V of 48-160 km/h. Each case was downloaded and summarized. Only investigator confirmed belted drivers were analyzed. Since most of the crash characteristics were similar to the 0° frontal fatalities, the data was combined. A subset of newer vehicles (≤ 5 years old) was compared to frontal NCAP, small overlap and OMDB (offset moving deformable barrier) crash tests. Photographs of the vehicle deformation, front crush and intrusion were compared. RESULTS: Thirty-three (33) fatal crashes met the selection criteria, adding to 16 in pure frontal impacts. The belted drivers were 53.7 ± 19.0 years old; 59.2% were male. The majority were White (61.9%). Alcohol was found in 25% of the drivers with 15% drug use. The vehicles were 11.7 ± 6.9 years old. The fatal injuries had MAIS of 4.2 ± 1.4. The chest (73.2%) was most frequently injured with multiple rib fractures, some with heart and aorta lacerations and hemothorax. The head (32.4%) was second with serious injuries to multiple body regions. The primary source of injury was the seatbelt and steering system (73.2%). The right-front interior (24.4%) was a source in +10, +20° PDOF crashes. The driver door and A-pillar (22.0%) were sources in -10, -20° PDOF crashes. The maximum frontal crush was 105.0 ± 26.7 cm with 55.1% involving occupant compartment intrusion. In 63.8% of the crashes, the steering wheel rim was deformed indicating heavy loading by the restrained driver. The delta V was 78.6 ± 19.2 km/h (48.8 ± 11.9 mph). Most (67.3%) of the crashes were on two lane roads with the case vehicle impacting another vehicle (77.6%) or drifting off road hitting a fixed object (22.4%). The most common frontal crash types were small overlap (34.7%), full width (28.6%), offset (20.4%) and oblique offset (12.2%). CONCLUSIONS: Fatal crashes with lap-shoulder belted drivers and PDOF up to ±20° of pure frontal involve vehicle deformations and delta Vs that are higher than tests conducted by NHTSA and IIHS. The driver deaths involve significant loading of the chest with multiple rib fractures and organ lacerations. If belted drivers are to be protected, the following need to be considered: higher and lower severity sled tests to optimize occupant restraints, innovations in vehicle structures to reduce intrusion and crash avoidance technologies to reduce the speed of fatal crashes. NHTSA's 90 km/h oblique OMDB test is not severe enough to bring about meaningful changes that lower traffic fatalities. A higher severity test should be considered.
Which factors influence rider injury severity in powered two-wheeler accidents? A systematic review.
Zhang Y, Zhang G, Xu J … +2 more , Li S, Liu Y
Traffic Inj Prev · 2026 Jan · PMID 41505729 · Publisher ↗
OBJECTIVES: The high accident risk associated with Powered Two-Wheelers (PTWs) and the limited generalizability of conclusions due to regional data disparities necessitate a comprehensive global analysis. This systematic...
OBJECTIVES: The high accident risk associated with Powered Two-Wheelers (PTWs) and the limited generalizability of conclusions due to regional data disparities necessitate a comprehensive global analysis. This systematic review aims to identify the influencing factors of injury severity in PTW accidents worldwide. METHODS: This study adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We conducted a comprehensive review of scientific literature on PTW accidents, ultimately identifying 32 high-quality peer-reviewed papers. The findings of these studies were thoroughly examined and synthesized using a six-dimensional classification framework (encompassing road, collision, vehicle, time, human, and environmental factors). RESULTS: The results indicated a consensus that the involvement of heavy vehicles, nighttime, weekends/holidays, summer, young/elderly riders, novice riders, drunk driving, frontal collisions, and running-off-the-road collisions significantly increase the severity of PTW accidents. Conversely, factors such as road classification, road geometry, intersections, weather conditions, and rider gender remain controversial, with the discrepancies likely stemming from the complex interplay of multiple variables. CONCLUSIONS: This systematic review integrates findings from diverse countries, offering valuable insights for global policymakers aiming to mitigate PTW accident severity. Furthermore, we identified methodological gaps in existing research and proposed future directions, thereby providing multiple avenues for researchers engaged in road safety studies.
Multimodal TOR interfaces for L3 autonomous vehicles: a tradeoff analysis of driver trust, cognitive load, and takeover efficiency.
Guo Q, Gong J, Yang Y
Traffic Inj Prev · 2026 Jan · PMID 41505719 · Publisher ↗
OBJECTIVES: The study focused on designing effective Takeover Request (TOR) interfaces to balance driver trust, cognitive load, and takeover efficiency in Level 3 automated driving. METHODS: This study used a driving sim...
OBJECTIVES: The study focused on designing effective Takeover Request (TOR) interfaces to balance driver trust, cognitive load, and takeover efficiency in Level 3 automated driving. METHODS: This study used a driving simulator to evaluate 16 multimodal interfaces that combined four visual designs, two auditory cues, and two haptic feedback states. Driving performance, subjective feelings, and eye-tracking metrics were measured, and a comprehensive ranking was established using the entropy weight method (EWM). RESULTS: The interface combining auditory and haptic cues with contextual visual information exhibited the best overall performance (EWM > 0.927). Auditory cues significantly improved driver trust (average improvement of 1.29 points, < .001) and enhanced takeover efficiency by reducing reaction time by 0.13 s ( < .001), demonstrating a 1.5 times the trust-building effect of haptic cues. Haptic cues effectively reduced cognitive load (DALI:Δ = -3.79, < .001) and compensated for processing delays in visually complex scenarios. For example, when combined with contextually enhanced visual information, it could reduce reaction time by 0.13 s ( < .001). CONCLUSIONS: Optimal TOR performance stems from strategic function allocation across modalities: utilizing auditory and haptic channels to issue emergency alerts while simultaneously using the visual channel for contextual information. This multimodal functional complementarity approach effectively balances trust, cognitive load, and efficiency, providing theoretical support for designing human-machine interfaces in next-generation autonomous vehicles.
Assessment of injury severity and recovery trajectories in urban motorcycle crashes: An analysis of hospital data from Dhaka, Bangladesh.
Ahmed N, Tasnim J, Hossain M … +5 more , Aftab KA, Kabir AKMF, Aktar F, Rahman AZMM, Nasim M
Traffic Inj Prev · 2026 Jan · PMID 41505706 · Publisher ↗
OBJECTIVE: Motorcycle crashes, one of the major contributors to worldwide road traffic crashes (RTCs), remain a critical public health issue often in several low- and middle-income countries, where data limitations hinde...
OBJECTIVE: Motorcycle crashes, one of the major contributors to worldwide road traffic crashes (RTCs), remain a critical public health issue often in several low- and middle-income countries, where data limitations hinder comprehensive safety assessments. Existing studies have largely relied on police-reported data, which often underrepresents crash severity and lacks detail on post-crash recovery. To address the gap, this study aims to utilize hospital-based data for examining the distribution of patients across demographic, vehicular, situational and crash dynamics related factors; and to provide a severity-based assessment across several variables, such as patient's age, impact speed, collision type, and impacted body region. Furthermore, it aims to evaluate severity-based post-crash outcomes, focusing on patient's 30-day recovery status. METHODS: The study used a questionnaire-based interview to collect data from 509 motorcycle crash victims admitted in two major hospitals, over a nine-month period in Dhaka, Bangladesh. The Abbreviated Injury Scale (AIS) was employed to provide a standardized assessment of injury severity, while the data summarization and visualization was achieved using R software package. RESULTS: Findings reveal a high frequency of motorcycle crashes among males in the 11-40 age bracket, middle-income individuals, and students or service-holders, with most incidents occurring at moderate speeds and head-on collisions on single-lane roads. Severity-based analysis highlights that injuries in AIS 2 (moderate) and 3 (serious) are predominant, with a trend of increasing proportion of injuries with severity of AIS 3 or above correlated with older age groups, high-speed crashes, and specific collision types like rear-end and head-on. Lower extremity, specially the legs, are the most susceptible body part to injuries across all severity categories. Moreover, recovery rates (proportion of injured patients who fully or partially recover from their injuries after 30-days) decrease with higher injury severity, with permanent disabilities being more common among severely injured patients. CONCLUSIONS: The data distribution reveals how crash frequency, and severity varies across several key factors while also divulging how injury severity impacts recovery trajectories. The results are expected to provide a foundation for future research and policy development toward reducing motorcycle crash risks and improving post-crash recoveries in similar urban settings.
Examining the effects of improved signs on lane-changing behavior at expressway emergency stop area.
Huang Y, Zhao Y, Jin Y … +4 more , Wang J, Xiang W, Li H, Li S
Traffic Inj Prev · 2026 Jan · PMID 41505608 · Publisher ↗
OBJECTIVE: The emergency stop area at expressways is designed to help drivers handle emergencies. However, to enter the emergency stop area, some risky lane-changing behaviors often occur. This study evaluated drivers' l...
OBJECTIVE: The emergency stop area at expressways is designed to help drivers handle emergencies. However, to enter the emergency stop area, some risky lane-changing behaviors often occur. This study evaluated drivers' lane-changing behavior under various sign schemes to identify an effective sign scheme that can improve lane-changing performance when entering the emergency stop area. METHODS: Based on the pre-position distance, disappearance distance of the signs, as well as the driver's judgment distance and action distance, different sign positions and number conditions were set. A driving simulator was used to examine drivers' lane-changing behavior when approaching the emergency stop area under various sign schemes. The lane-changing behavior was divided into the decision process and the action process. The Generalized Estimating Equations (GEE) model was used to analyze five selected behavioral variables. Besides, the entropy-weighted TOPSIS model was employed to select the recommended sign scheme. RESULTS: The results show that both the Single position improved (SPI) sign and the Two-stage improved (TSI) sign effectively advanced the drivers' lane-changing position by over 52.4%, increasing the lane-change success rate by more than 23.8% compared to the baseline sign. The TSI sign further increased the lane-changing speed and improved comfort compared to the SPI sign. The entropy-weighted TOPSIS model results identified TSI380 as the recommended sign scheme. Additionally, mature drivers performed better in lane-changing performance than novice drivers, and female drivers were more likely to follow the sign information. CONCLUSIONS: The findings provide scientific guidance for the placement of signs before emergency stop areas. It is recommended to set signs at 380 m and 230 m before the entrance of the emergency stop area to reduce risky lane-changing behavior.
Challenges for older drivers: Workload and risky behavior in simulated six-lane urban bidirectional road conflict scenarios.
Zhou Y, Guo F, Wang J … +2 more , Cai J, Wan H
Traffic Inj Prev · 2026 Jan · PMID 41503822 · Publisher ↗
OBJECTIVE: This study aims to investigate older drivers' driving workload, risky driving behavior characteristics, and their interrelationships. It further examines how driving workload, demographic characteristics, and...
OBJECTIVE: This study aims to investigate older drivers' driving workload, risky driving behavior characteristics, and their interrelationships. It further examines how driving workload, demographic characteristics, and driving fitness influence risky driving behaviors, providing a theoretical foundation for the safety management of older drivers. METHODS: A typical bidirectional six-lane urban road scenario was constructed to conduct a driving simulation experiment involving various conflict situations under both left-turn and right-turn conditions. Behavioral data of older drivers were collected to analyze and evaluate their driving fitness, driving workload, and risky driving behavior performance. A generalized linear mixed model was employed to examine the underlying mechanisms between the different variables and risky driving behaviors. RESULTS: The findings indicated that older drivers exhibited higher operational intensity in left-turn scenarios, whereas their risky driving behavior indicators were generally lower in pedestrian conflict situations. The workload during motorcycle and pedestrian conflict scenarios was significantly higher than that in car conflict situations. Moreover, the self-reported workload in left-turn maneuvers was significantly higher than in right turns. GLMM analysis revealed that the mental demand dimension of workload significantly affected risky driving behaviors. Furthermore, longer driving mileage and better driving fitness were significantly associated with fewer risky driving behaviors. CONCLUSION: The findings provide theoretical implications for older driver safety management and underscore the importance of incorporating driving fitness assessment and subjective workload monitoring into comprehensive safety management frameworks.
Patterns and predictors of co-occurring risky behaviors among motorcyclists.
Zhao X, Lee C, Wright S
Traffic Inj Prev · 2026 Jan · PMID 41503803 · Publisher ↗
OBJECTIVES: Motorcyclists, as a vulnerable group of road users, not only suffer from trauma caused by collisions with other vehicles but also from the consequences of their own risky behaviors. As the number of motorcycl...
OBJECTIVES: Motorcyclists, as a vulnerable group of road users, not only suffer from trauma caused by collisions with other vehicles but also from the consequences of their own risky behaviors. As the number of motorcycle crashes has continued to rise in recent years, it is essential to understand trends in motorcyclists' risky behaviors to enhance their safety. This study investigates the co-occurrence of six risky riding behaviors-speeding, alcohol-impaired riding, helmet nonuse, distraction, fatigue, and anger-and identifies rider characteristics associated with distinct behavioral profiles. METHODS: This study analyzes data from 4,390 motorcyclists who participated in three annual surveys conducted in Florida between 2021 and 2023. The survey collected information on demographics, riding habits, risky motorcycling behaviors, and the perception that other drivers fail to notice or attend to motorcycles. Descriptive statistics and latent class analysis (LCA) were conducted to identify subgroups of riders based on co-occurring behaviors. A bias-adjusted three-step approach was used to examine associations between class membership and rider characteristics. RESULTS: Four latent behavioral classes were identified: Very High-risk Riders, Moderate-Risk Riders, Drinking Riders, and Low-risk Riders. Each group exhibited distinct constellations of risky behaviors and demographic or attitudinal profiles. High-risk riders were more likely to be younger, less trained, and not using any safety gear. Alcohol use was especially pronounced among recreational riders regardless of age or gender. Formal motorcycle training was consistently linked to lower-risk classes. CONCLUSION: This study advances the understanding of motorcyclist safety by demonstrating how risky behaviors cluster and how specific demographic and perceptual factors distinguish rider subgroups. These findings support the development of comprehensive safety interventions that target subgroups of motorcyclists who share similar risky profiles, rather than treating isolated behaviors.
Antemortem toxicological and behavioral findings of drivers involved in fatal traffic crashes: a comparative analysis of charged and uncharged individuals.
Lee D, Corral SR, Mance A … +1 more , Stout P
Traffic Inj Prev · 2026 Jan · PMID 41503796 · Publisher ↗
OBJECTIVES: Alcohol/drug-related impairment and unsafe driving behaviors such as speeding, running red lights, or failure to wear a seat belt have been identified as contributing factors to road traffic accidents. This s...
OBJECTIVES: Alcohol/drug-related impairment and unsafe driving behaviors such as speeding, running red lights, or failure to wear a seat belt have been identified as contributing factors to road traffic accidents. This study aims to examine antemortem toxicological testing results, driving behaviors and other characteristics of drivers involved in fatal traffic crashes, which may offer insights into the broader spectrum of fatal crash risk factors. METHODS: Toxicological results and case data including demographic characteristics and driving behaviors were analyzed for drivers charged in fatal traffic crashes and thus potentially culpable between 2019 and 2024. These individuals were compared with drivers involved in fatal collisions who had not been charged. Whether a driver was charged with an accident depended not only on culpability but also on the availability of sufficient evidence, the presence of criminal negligence, legal/procedural considerations, contributory faults, and other contextual factors. All drivers survived the accidents and provided antemortem blood samples, which were submitted by the Houston Police Department (HPD) for alcohol and drug testing to the Houston Forensic Science Center (HFSC). RESULTS: Based on offense classifications in the HFSC database, HPD reports, and court records, 116 drivers were identified as suspects charged in fatal crashes in 2019-2024. Among the charged drivers, 81% were male with an average age of 32 years (range: 17-62), and 97% tested positive for alcohol and/or drugs. Ethanol was the most frequently detected substance (73%), followed by cannabinoids (47%) and cocaine/metabolites (12%). Common unsafe driving behaviors included speeding/failure to control speed (53%), lane deviation (31%), and running red lights/stop signs (20%). In comparison, of 87 uncharged drivers (82% male, average age 40 years) involved in fatal crashes, 61% tested negative for both alcohol and drugs. Among the 39% who tested positive, cannabinoids (17%) and ethanol (9%) were the most frequently detected drugs. The majority (91%) of the charged drivers committed one or more traffic infractions. Failure to control speed/speeding was the most common (53%), followed by failure to drive in a single lane (31%). In contrast, the majority (64%) of the uncharged drivers did not commit a traffic infraction. CONCLUSIONS: The present research revealed several risk factors of fatal traffic crashes - younger male drivers with high blood alcohol concentrations, drug-impaired drivers who failed to control speed or to drive in a single lane, pedestrians who did not yield the right of way to vehicle, and motorcyclists who sped. Targeting these risk factors could aid in addressing the increasing number of traffic fatalities.
Multi-section landscape intervention for driving fatigue in extra-long tunnels: A real-road driving study based on visual perception and adaptation.
Ding X, Du Z, He S … +1 more , Wu P
Traffic Inj Prev · 2025 Dec · PMID 41417460 · Publisher ↗
OBJECTIVE: The weak visual reference environment and prolonged driving duration in extra-long tunnels may induce driving fatigue accumulation, potentially compromising traffic safety. This study aims to investigate the i...
OBJECTIVE: The weak visual reference environment and prolonged driving duration in extra-long tunnels may induce driving fatigue accumulation, potentially compromising traffic safety. This study aims to investigate the impact mechanisms of multi-section landscape design on drivers' visual perception and adaptation capacities, and evaluate its effectiveness in alleviating driving fatigue. METHOD: This study was conducted in the Qinling Zhongnanshan Extra-Long Tunnel, which contained three special landscape zones. A total of thirty-two drivers were recruited to participate in real-world driving tests, and their eye movement behavior and pupil area data were collected using an eye tracker. The drivers' field of view was divided into six areas of interest (AOIs), and the spatial transfer characteristics of gaze points were analyzed. By combining kernel density estimation and visual sensitivity area (VSA) analysis, the differences in drivers' visual search breadth between the ordinary and fatigue-alleviating sections were compared. Additionally, the relative change rate of pupil area (RCRP) indicator was established to quantitatively evaluate the drivers' visual adaptation process to the landscape design. RESULTS: Landscape zones significantly improved drivers' visual perception patterns, with increased gaze transfer probabilities to roadside and top areas, and expanded horizontal VSA ranges. However, a noticeable contraction in gaze range occurred in the rear section of the tunnel, accompanied by a significantly decreased gaze transfer frequency. Pupil change analysis indicated that the landscape stimuli did not cause visual discomfort (RCRP < 20%), but revealed more sensitive pupil responses in middle and rear tunnel sections, suggesting a gradual decline in visual adaptation ability. CONCLUSION: The study demonstrates that multi-section landscape zones effectively regulate the allocation of drivers' visual attention by reconstructing the visual environment within tunnels. However, the intervention efficacy progressively attenuates with fatigue accumulation. Notably, drivers' visual adaptation capacity deteriorates prior to measurable declines in environmental perception. These findings provide critical evidence for optimizing fatigue-alleviating landscape design in extra-long tunnels.
Determinants of following distance and fatality risk in motorcycle rear-end collisions in Thailand.
Prajongkha P, Kanitpong K
Traffic Inj Prev · 2025 Dec · PMID 41417257 · Publisher ↗
OBJECTIVE: Critical following distance, particularly in motorcycle-following behavior, contributes to rear-end collisions that result in severe injuries and fatalities for motorcyclists. This study investigates the key f...
OBJECTIVE: Critical following distance, particularly in motorcycle-following behavior, contributes to rear-end collisions that result in severe injuries and fatalities for motorcyclists. This study investigates the key factors affecting motorcycle following distance within the critical following distance and the fatality probability of rear-end collisions due to insufficient following gaps. Naturalistic traffic data and real motorcycle rear-end crashes from 2017 to 2021 were analyzed. METHOD: The critical following distance concept is proposed using time headway and time-to-collision derived from trajectory traffic data (Prajongkha and Kanitpong 2024). Binary logistic regression is employed to analyze factors contributing to critical following distance and to assess fatality probability in real motorcycle rear-end collisions. The outcomes of the regression models are verified through likelihood ratio tests and Nagelkerke R-squared values. RESULT: Speed, relative speed, helmet use, a single leading vehicle, and vehicle type significantly influence entrance into the critical following distance. Motorcyclists tend to maintain a shorter longitudinal gap when following a car than when following a motorcycle or truck. Helmet use reflects cautious behavior, as helmeted motorcyclists maintain larger gaps. In real motorcycle rear-end collisions, pre-impact speed and leading vehicle type-especially a truck-significantly increase fatality probability. An in-depth analysis reveals that emergency reactions involving only braking, without evasive action, are more likely to result in fatalities, accounting for 7 out of 10 fatal cases. CONCLUSION: Motorcyclists who do not wear helmets exhibit a higher likelihood of approaching critical following distance. Regarding fatality probability, high-speed impact and collisions with larger vehicles, such as trucks, are the primary contributing factors. The critical following distance concept can serve as a valuable reminder system for motorcyclists to maintain safer following gaps.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe