Traffic Inj Prev [JOURNAL]
Sun
200 papers
Preparations reduce drivers' perceived risk when resuming control from conditionally automated vehicles.
Pan H, Stephens A, Zhao Z … +3 more , Huo Y, Liu Y, Koppel S
Traffic Inj Prev · 2026 Jun · PMID 42296445 · Publisher ↗
OBJECTIVE: During conditionally automated driving, an unexpected vehicle control transfer increases a driver's perceived risk and negatively impacts their subsequent driving behavior. Monitoring requests (MRs) that occur...
OBJECTIVE: During conditionally automated driving, an unexpected vehicle control transfer increases a driver's perceived risk and negatively impacts their subsequent driving behavior. Monitoring requests (MRs) that occur before takeover requests (TORs) have been shown to enhance takeover performance. However, whether this also influences a driver's perceived risk is not clear. This study aimed to examine the effect of MRs on drivers' perceived risk, investigate whether their physiological indicators can predict perceived risk, and explore the association between perceived risk and takeover performance. METHOD: Sixty participants (mean age = 28.1, = 5.0; gender = 46.7% men, 53.3% women) were randomly distributed into two groups and completed a simulated automated driving experiment. One group experienced MRs, followed by TORs (MRs + TORs), while the other group experienced TORs only (TORs-only). Both groups experienced two takeover scenarios during which their perceived risk, physiological indicators (including eye-movement-related indicators, heart rate, skin conductance level, and respiration frequency) and takeover performance (including maximal braking pedal input [0-1] and maximal steering wheel velocity [rad/s]) were collected. Additionally, participants' self-reported perceived risk, on an 11-point Likert scale, regarding the takeover scenarios was also recorded. A random forest model was applied to predict perceived risk using drivers' pre- and post-TOR physiological indicators as input features. RESULTS: Compared to drivers in the TORs-only group, drivers in the MRs + TORs group reported a lower perceived risk, exhibited a higher average amplitude of saccades before the TORs, along with lower indicator variations after the TORs, as well as lower maximal braking pedal input and maximal steering wheel velocity. The perceived risk prediction model showed good performance with a mean macro-precision of 0.757, a mean balanced accuracy of 0.772, and a mean macro-F1 of 0.758. CONCLUSIONS: This research found that MRs before TORs were associated with reduced drivers' perceived risk, and that pre- and post-TOR physiological indicators showed potential for predicting perceived risk. These findings would inform the development of driver-state detection techniques and the design of human-machine interaction strategies.
How do driving speed and luminance change rate affect driving visual comfort? A field vehicle test on mountain highway tunnels.
Meng Y, Huang T, Ren C … +2 more , Li B, Xia Y
Traffic Inj Prev · 2026 Jun · PMID 42296444 · Publisher ↗
OBJECTIVE: This study aims to investigate the factors influencing drivers' visual-psychological comfort at the tunnel portal sections of two-lane mountainous highways, as well as the mechanism by which the environmental...
OBJECTIVE: This study aims to investigate the factors influencing drivers' visual-psychological comfort at the tunnel portal sections of two-lane mountainous highways, as well as the mechanism by which the environmental conditions at these sections affect drivers' visual-psychological comfort. The findings are intended to provide a theoretical basis for the safety design and traffic management of tunnel portal sections. METHODS: Data were collected through field vehicle tests. The Principal Component Analysis (PCA) was employed to fuze the driving visual and psychological indicators. The K-means clustering algorithm was adopted to classify comfort levels into three categories: "Comfortable", "Relatively Comfortable", and "Uncomfortable". The ordered logistic regression method was utilized to analyze the impacts of eight variables on comfort levels. Furthermore, the decision tree model was applied to identify the key nodes and threshold values of indicators that lead to the differences in comfort levels. RESULTS: Driving experience, driving speed, luminance change rate, and driver temperament type exert a significant influence on visual-psychological comfort, among which driving speed is the dominant factor and luminance change rate is the secondary factor. A driving speed of 47 km/h acts as the threshold differentiating the "comfortable" and "relatively comfortable" states, whereas 53 km/h marks the threshold between "relatively comfortable" and "uncomfortable". Similarly, a luminance change rate of 8.5% is the threshold for distinguishing "comfortable" from "relatively comfortable", and 28.5% is the threshold separating "relatively comfortable" from "uncomfortable". CONCLUSIONS: Driving speed and luminance change rate are the two most critical controllable environmental factors that affect driving visual psychological comfort. The study results offer specific quantitative thresholds for speed limit setting and lighting transition design of these road sections, which may help enhance driving safety and improve driving experience.
Drivers' visual fixation in spiral tunnels: Impacts of geometric parameters and travel direction on cognitive load.
Han L, Zhou H, Gu P … +1 more , Du Z
Traffic Inj Prev · 2026 Jun · PMID 42296442 · Publisher ↗
OBJECTIVE: This study aims to systematically investigate how key geometric parameters of spiral tunnels, specifically tunnel length and curve radius, interact with travel direction, including uphill and downhill, to infl...
OBJECTIVE: This study aims to systematically investigate how key geometric parameters of spiral tunnels, specifically tunnel length and curve radius, interact with travel direction, including uphill and downhill, to influence drivers' visual fixation behavior and associated cognitive load, and to reveal the underlying visual adaptation mechanisms in curved and enclosed tunnel environments. METHODS: A real vehicle naturalistic driving experiment was conducted in three operational highway spiral tunnels with distinct geometric configurations: Nanping (1,330 m length, 1,000 m curve radius), Liuyuan (2,200 m length, 850 m curve radius), and Hankou (4,460 m length, 700 m curve radius). Eye movement data from 30 licensed drivers were collected using a wearable eye tracker. Four fixation-related metrics, including fixation duration, fixation frequency, horizontal fixation deviation, and vertical fixation deviation, were analyzed using two-way ANOVA and post hoc tests. RESULTS: Tunnel geometric parameters exerted significant main effects on all fixation metrics. With increasing tunnel length and decreasing curve radius, fixation duration increased monotonically. For uphill driving, it rose from 490.28 ms to 552.18 ms. Fixation frequency and horizontal fixation deviation decreased significantly. Uphill frequency dropped from 2.85 Hz to 2.05 Hz, and horizontal deviation decreased from 21.64 degrees to 12.09 degrees. In contrast, vertical fixation deviation increased progressively from 13.70 degrees to 21.06 degrees for uphill driving. Travel direction also showed significant main effects. Uphill traversal consistently produced longer fixation durations, higher fixation frequencies, and broader horizontal and vertical deviations than downhill traversal under identical geometric conditions. CONCLUSIONS: Spiral tunnel geometry induces a dual visual adaptation mechanism: horizontal visual tunneling coupled with compensatory vertical scanning for far-field path preview. Uphill traversal imposes higher cognitive load than downhill traversal, revealing a directional asymmetry in driving demands. These findings provide empirical evidence for human-centered spiral tunnel design and offer quantitative benchmarks for driver monitoring and adaptive assistance systems in tunnel scenarios.
A deep clustering spatiotemporal framework for real-time urban traffic risk probability estimation.
Luo Y, Ranjitkar P, Xu J … +4 more , Xin Y, Li Y, Liu C, Li Y
Traffic Inj Prev · 2026 Jun · PMID 42296441 · Publisher ↗
OBJECTIVES: Urban road traffic exhibits complex and highly dynamic flow patterns, making real-time risk probability assessment challenging. Existing measures such as Time to Collision (TTC) rely on full sample vehicle tr...
OBJECTIVES: Urban road traffic exhibits complex and highly dynamic flow patterns, making real-time risk probability assessment challenging. Existing measures such as Time to Collision (TTC) rely on full sample vehicle trajectory data, which are difficult to obtain at large urban scales. To address this limitation, this study adopts a spatiotemporal grid representation and utilizes floating vehicle trajectory data to characterize traffic operational states and develop a real-time urban traffic risk probability assessment framework. METHODS: This study proposes an integrated spatiotemporal framework for urban traffic risk probability estimation. A Graph Attention Network Long Short Term Memory (GAT-LSTM)-based Spatiotemporal Autoencoder Neural Network (GL-SANN) is developed to extract latent risk features from vehicle operational parameters. These features are further incorporated into a Deep Clustering Spatiotemporal Network (DCSN) with K-means clustering to assess traffic risk levels. A LightGBM model is then constructed for real-time risk identification. Finally, a Spatiotemporal Graph Convolutional Risk Prediction (SGCRP) model is designed to predict future vehicle operational parameters and infer short-term risk states. RESULTS: Experiments using ride-hailing vehicle trajectory data from Xi'an demonstrate that traffic risk probability patterns can be classified into five levels, with risk probability levels increasing significantly during peak periods and decreasing during off-peak periods. Specifically, the proportion of high-risk grids reaches approximately 50%-60% during peak periods, while during off-peak periods, this proportion decreases to about 37%. Intersections and traffic-intensive grids consistently exhibit higher risk probability levels than ordinary grids. The DCSN model consistently outperforms benchmark methods. The LightGBM-based risk identification model achieves a precision of 0.984. The proposed risk prediction model improves training speed by 33.1% over the Transformer model and yields a prediction precision of 0.974. CONCLUSIONS: These findings provide a method for proactive urban traffic risk prevention and contribute to the development of intelligent transportation systems.
A multi-state methodological framework for behavioral surveillance of cannabis-impaired driving.
Ageze D, Baird S, Hacker S … +6 more , Dell'Acqua R, Gold A, Shaughnessy T, Lanin-Kettering I, Marcotte TD, Hill LL
Traffic Inj Prev · 2026 Jun · PMID 42275458 · Publisher ↗
OBJECTIVES: Driving under the influence of cannabis is a growing public health concern in the United States. Existing surveillance systems often lack behavioral detail, rely on static quotas, or fail to account for diffe...
OBJECTIVES: Driving under the influence of cannabis is a growing public health concern in the United States. Existing surveillance systems often lack behavioral detail, rely on static quotas, or fail to account for differences across legal contexts. The Cannabis and Roadway Safety Study (CARSS) is a multi-phase project designed to address these gaps. This paper describes the methods of the Phase 2 questionnaire. METHODS: A cross-sectional, mixed-methods questionnaire assessed cannabis use patterns, driving behaviors, impairment beliefs, and messaging preferences among adult drivers who currently use cannabis. The study employed a structured online questionnaire incorporating quantitative items and AI-assisted qualitative probes. Screening, quota sampling, and rake weighting were used to achieve state-level demographic representation based on 2020 U.S. Census benchmarks, with real-time monitoring. A secondary weighting step aligned full questionnaire participants with adult drivers who currently use cannabis within each state. RESULTS: The framework was implemented across eight U.S. states representing non-medical, medical-only, and prohibition policy environments, with data collection from November 28, 2023, to February 13, 2024. All states achieved target sample sizes of approximately 250 respondents, yielding a total analytic sample of 2,023 participants. Recruitment dynamics varied by state, with some requiring extended fielding to meet demographic targets. Usable open-text responses were obtained in all states through AI-assisted probing across policy environments. CONCLUSIONS: CARSS Phase 2 demonstrates the feasibility of integrating responsive quota monitoring, multi-step weighting, and AI-assisted qualitative probing into a scalable, multi-state surveillance framework for cannabis-impaired driving research and related public health.
Estimation of optimum speed limit on urban expressway: case study of Delhi-Meerut and Greater Noida Expressways.
Verma G, Chalumuri RS, Chandra S
Traffic Inj Prev · 2026 Jun · PMID 42275044 · Publisher ↗
OBJECTIVE: This study introduces a data-based approach for determining the speed limits on urban expressways, using the Delhi-Meerut and Greater Noida Expressways as case study. It also aims to evaluate the relationship...
OBJECTIVE: This study introduces a data-based approach for determining the speed limits on urban expressways, using the Delhi-Meerut and Greater Noida Expressways as case study. It also aims to evaluate the relationship between speed behavior characteristics and crash frequency, and to identify optimal speed limits that can enhance road safety while maintaining traffic efficiency. METHODS: An examination is conducted utilizing three years of past crash data combined with speed percentile matrices and traffic infraction data. Four modeling techniques Linear Regression (LR), Non Linear Regression (NLR), Artificial Neural Network (ANN), and Extreme Gradient Boosting (XGBoost) are utilized to estimate crash frequency based on key speed behavior parameters like the 85th, 50th, and 15th percentile speeds, speed variability, and rates of violations. The models are assessed using standard performance metrics such as , test, and test. RESULTS: The XGBoost model demonstrates better predictive accuracy; nevertheless, the NLR model is favored for its clarity and simplicity in reverse calculations to determine the ideal speed limits. Utilizing the NLR-based model, crash forecasting is employed to establish acceptable thresholds for crash rates, and suitable optimal speed limits are suggested for expressway sections. Optimization findings indicate that the existing speed limits for two Expressways, 70 km/h for cars and 50 km/h for heavy vehicles and 75 km/h for car and 50 km/h for Heavy vehicle surpass the safe levels, optimal limits are projected to be 60 km/h for cars and 40 km/h for heavy vehicles on Delhi-Meerut Expressway and 70 km/h for cars and 45 km/h for Heavy Vehicle on Greater Noida Expressway. Moreover, variable speed limits influenced by time-of-day patterns were suggested, advising 65 km/h during off-peak times and 55 km/h during the peak traffic hours. CONCLUSIONS: This study provides an empirical basis to bolster speed management strategies focused on improving road safety in high-speed urban routes in India. The findings highlight the importance of adopting data-driven and dynamic speed limit policies that consider real-time traffic conditions and speed behavior characteristics. The approach can be extended to other urban expressways with similar traffic conditions, contributing to the development of safer and more efficient transportation systems.
Validation and application of automated CT analysis to musculoskeletal profiling in MVC occupants.
Madrid Fuentes DA, Armstrong W, Hsu FC … +13 more , Bowles J, Guerrero Sanchez GB, Popuri K, Beg MF, May AK, Torres Fajardo RA, Lopez R, Kiani B, Martin RS, Miller AN, Devane K, Stitzel JD, Weaver AA
Traffic Inj Prev · 2026 Jun · PMID 42275029 · Publisher ↗
OBJECTIVE: Bone mineral density (BMD) and muscle health influence an occupant's tolerance to motor vehicle crash (MVC), and these musculoskeletal factors can be used to design more effective countermeasures. CT imaging r...
OBJECTIVE: Bone mineral density (BMD) and muscle health influence an occupant's tolerance to motor vehicle crash (MVC), and these musculoskeletal factors can be used to design more effective countermeasures. CT imaging routinely acquired during trauma evaluation can be leveraged to obtain volumetric (v)BMD and muscle metrics, yet extracting measurements manually from CT is time-consuming and inconsistent, typically limiting it to single anatomical regions. In this retrospective observational study, we applied an existing AI-based CT segmentation tool to characterize the musculoskeletal status of MVC occupants. METHODS: An external cohort of 55 adults with CT scans including a bone calibration phantom was used to validate the automated segmentation of trunk muscle cross-sectional area (CSA) and an automated tissue-calibrated method for vBMD. Agreement with manual CSA and phantom-calibrated vBMD was assessed via correlation and Bland-Altman analyses. The validated automated pipeline was then applied to 851 CT scans from Crash Injury Research and Engineering Network (CIREN) cases collected between 2005-2023. The Data Analysis and Facilitation Suite (v3.11.3, Voronoi Health Analytics) software was used to measure lumbar spine (L1-L4), pelvis, femur head, femur neck and femur trochanter + shaft vBMD, as well as trunk muscle CSA and density. Sarcopenia was defined as a trunk muscle CSA divided by height squared <38.5 cm/m for females and <52.4 cm/m for males. Osteopenia was defined as lumbar spine vBMD <145 mg/cm. These characteristics were examined in association to regional fracture count with abbreviated injury scale (AIS) severity and injury severity score (ISS) using negative binominal regression, including crash variables (delta-V, belt status, driver status, airbag deployment, principal direction of force, model year, curb weight) and occupant characteristics (age, sex, height, weight). RESULTS: In the validation cohort (64% female; ages 66 ± 4), automated and manual trunk muscle CSA measurements agreed ( = 0.996; < 0.0001), with small average differences (-4.9 cm). Automated tissue-calibrated lumbar vBMD closely matched phantom-calibrated values ( = 0.938, mean difference -0.21 mg/cm³), and femur regions showed similar agreement (correlation = 0.865-0.927; all < 0.0001).In CIREN occupants (56% female; ages 47 ± 20), 31% had sarcopenia, 28% had osteopenia, and 14% osteosarcopenia according to the CT-based definitions. vBMD declined with age across all regions (all < 0.0001). Lower vBMD was associated with higher ISS in the pelvis ( = 0.003), femur head ( < 0.0001), femur neck ( < 0.0001), and trochanter + shaft ( = 0.00087). Additionally, a 10 mg/cm³ higher femur head vBMD was associated with 1.6% fewer AIS 2 fractures (-value = 0.005). CONCLUSIONS: Application of an AI-based segmentation platform to MVC CT scans demonstrated that compromised musculoskeletal tissue quality is common and associated with fractures and injury severity beyond traditional occupant and crash characteristics. This suggests that musculoskeletal profiling can inform MVC injury prevention, occupant protection design, and post-crash care strategies.
Analysis of real word safety benefits of pedestrian autonomous emergency braking (PAEB) for pedestrians, cyclists and motorcyclists.
Keall MD, Newstead S
Traffic Inj Prev · 2026 Jun · PMID 42275024 · Publisher ↗
OBJECTIVE: Autonomous Emergency Braking (AEB) fitted to vehicles has proven road safety benefits in terms of preventing collisions with other vehicles. AEB with pedestrian detection (PAEB) extends this capability to prot...
OBJECTIVE: Autonomous Emergency Braking (AEB) fitted to vehicles has proven road safety benefits in terms of preventing collisions with other vehicles. AEB with pedestrian detection (PAEB) extends this capability to protect vulnerable road users: the current paper evaluates these safety benefits in comparison with AEB without pedestrian detection and vehicles without any AEB. METHODS: We analyzed data from Australian and New Zealand crashes involving injury over the period 2016-2023. The target crash types for PAEB analyzed either involved a vulnerable road user or the vehicle was rear-ended by another vehicle. Vehicles listed in Redbook (which provides vehicle specification information to assist Australian vehicle purchasers) with the feature "Control - Pedestrian Avoidance with Braking" were classed as having PAEB; for other vehicles, "Collision Mitigation - Forward (High speed)" and "Collision Mitigation - Forward (Low speed)" identified other forms of AEB. Using an induced exposure approach, crash rates associated with the safety technology fitted were estimated. RESULTS: Controlling for jurisdiction, speed limit area, vehicle market group, driver age group and sex, weather, day/night and year of crash, there was a 17% reduction (95% CI 5% to 27%) in the rate of pedestrian or cyclist collisions associated with vehicles equipped with PAEB relative to vehicles without any AEB system; for vehicles equipped with AEB without pedestrian detection, there was no reduction in the rate of pedestrian or cyclist collisions. For collisions with motorcycles, the associated benefits were smaller: compared to vehicles with no AEB system, there was a statistically significant 11% reduction associated with PAEB (95% CI 0% to 21%) and a non-significant 8% reduction (95% CI an increase of 4% to a decrease of 19%) associated with other AEB system fitment. There were insufficient cyclists in the data to estimate benefits specifically for cyclists but there was no evidence that the associated benefit was any less than for pedestrians. The safety benefits for vulnerable road users associated with AEB with pedestrian detection were estimated to be higher at nighttime, contrary to some findings in other studies. CONCLUSIONS: The safety benefits associated with PAEB for vulnerable road users are significantly greater than AEB without pedestrian detection, and this analysis of real-world crash outcomes shows these benefits extend to the protection of motorcyclists.
Risk factors and outcomes of blunt cardiac injury in adult motor vehicle collision patients.
Shahbaz H, Sherman AB, Shaikh FA … +4 more , Elsawwah JK, Charles EJ, Curran T, Nemeth ZH
Traffic Inj Prev · 2026 Jun · PMID 42274641 · Publisher ↗
OBJECTIVES: Motor vehicle protective equipment, such as seatbelts and airbags, has improved occupant safety. However, while seatbelts reduce facial and abdominal injuries, they may not significantly prevent head, neck, o...
OBJECTIVES: Motor vehicle protective equipment, such as seatbelts and airbags, has improved occupant safety. However, while seatbelts reduce facial and abdominal injuries, they may not significantly prevent head, neck, or thoracic trauma. Limited data exist on blunt cardiac injury (BCI). This study evaluated patterns of BCI, associated thoracic injuries, and hospital outcomes in adult trauma patients following motor vehicle collisions (MVCs). METHODS: We analyzed the 2023 American College of Surgeons Trauma Quality Improvement Program (ACS-TQIP) database for adult MVC occupants. Abbreviated Injury Scale codes 4208xx.x, 4404xx.x, 4410xx.x, 4412xx.x, 4413xx.x, and 4416xx.x identified patients with BCI. Those without BCI formed the reference cohort. A 1:1 propensity score match (PSM) on Injury Severity Score (ISS) was performed using RStudio to balance collision severity. RESULTS: In the overall cohort, the incidence of BCI was 1.2% (1,914/161,446). After PSM, 1,914 patients remained in each cohort with a mean ISS of 22.7. Both seatbelt plus airbag use and airbag use alone were independently associated with increased odds of BCI. BCI was strongly associated with thoracic injuries, including sternum fracture (odds ratio [OR] 3.492; 95% CI 2.95-4.14), hemothorax (OR 2.928; 95% CI 2.29-3.75), thoracic aortic injury (OR 1.773; 95% CI 1.29-2.44), and pulmonary contusion (OR 1.382; 95% CI 1.18-1.62). In multivariable analysis with BCI as the outcome, mortality (OR 2.325; 95% CI 1.93-2.79) and cardiac arrest (OR 1.827; 95% CI 1.29-2.59) were independently associated with BCI. CONCLUSION: Protective equipment use correlates with BCI and thoracic trauma. In MVC patients using seatbelts and airbags, concomitant chest injuries should heighten suspicion for BCI and prompt further evaluation.
Head acceleration event frequency among drivers across track types in Stock Car Auto Racing.
Cavallero NJ, Mink ZR, Pritchard NS … +5 more , Bullock GS, Harper MG, Urban JE, Patalak JP, Stitzel JD
Traffic Inj Prev · 2026 Jun · PMID 42246232 · Publisher ↗
OBJECTIVES: Repetitive, non-concussive head acceleration event (HAE) frequency has not been characterized across track types in professional stock car racing. The objective of this work was to calculate and compare HAE f...
OBJECTIVES: Repetitive, non-concussive head acceleration event (HAE) frequency has not been characterized across track types in professional stock car racing. The objective of this work was to calculate and compare HAE frequency among drivers across track types in the National Association for Stock Car Auto Racing (NASCAR). METHODS: Head kinematics for 61 drivers were monitored over 39 race weekends at 30 distinct tracks during the 2025 season. Drivers across NASCAR's National series (Truck, Xfinity, and Cup) were instrumented with a custom mouthpiece sensor that tightly couples to the maxillary dentition containing a tri-axial linear accelerometer and gyroscope. A HAE was recorded when a 4 g trigger threshold was exceeded on any axis for at least 3 ms at 1600 Hz. HAEs were identified through filters (time-windowing, wear status) for each driver in each race where they participated and data was collected successfully. Events were visually verified and classified as crash events when associated with caution-causing incidents during active racing; all remaining real events were classified as race events. The number of race events for each driver was then divided by the number of laps completed by that driver to determine HAE rate (HAEs/lap). HAE rates were also calculated on a per-hour and per-race basis using transponder timing data and the number of races where the device was worn and data successfully collected. For this analysis, only feature race HAEs were included due to consistent session type and event data collection. Rates were aggregated to determine summary statistics of mean, median, 95 percentile, and maximum HAE frequencies. Rates were also compared using negative binomial regression to determine rates ratios and 95% confidence intervals. RESULTS: A total of 5,131 HAEs were collected over 239,337 feature racing laps. The lowest number of races ( = 14), laps completed ( = 14,752), and driver-races ( = 195) occurred at road courses with the greatest occurring at intermediate tracks ( = 44, = 132,419, = 621, respectively). Road courses resulted in significantly greater rates ( < 0.05) and the highest mean, median, 95 percentile, and maximum rates compared to all track types, where per-lap rates were 0.1246, 0.0733, 0.4264, and 0.5565 HAEs/lap. Comparatively, superspeedways had the lowest mean, median, and maximum rates at 0.0046, 0.0027, and 0.0222 HAEs/lap. The lowest 95%ile HAEs/lap occurred at intermediate tracks (0.0172). CONCLUSIONS: Instrumented mouthpieces were used to quantify HAE frequency across track types. Calculating HAE frequency on a per lap basis allows for comparisons of drivers across track types. Road courses had the greatest, mean, median, 95%ile, and maximum HAE rates. This study provides empirical evidence that warrants further investigation into road courses, which may elucidate factors contributing to increased rates. These findings additionally provide a framework for assessing race HAE rates that could provide insights to inform evidence-based decisions to improve driver safety.
Subarea-based modeling of crash severity in freeway ramp zones: Exploring heterogeneity and transferability.
Pan Y, Li P
Traffic Inj Prev · 2026 Jun · PMID 42246230 · Publisher ↗
OBJECTIVES: The frequent acceleration, deceleration and lane change of vehicles in the freeway ramp area lead to significant differences in the accident mechanism of the diverging, merging, and weaving zones. This study...
OBJECTIVES: The frequent acceleration, deceleration and lane change of vehicles in the freeway ramp area lead to significant differences in the accident mechanism of the diverging, merging, and weaving zones. This study aims to analyze the heterogeneity and transferability of factors affecting the severity of accident injuries in different ramp sub-areas, and to make a comparative analysis of the significant influencing factors in different sub-areas. METHODS: Based on the accident data of freeway ramps in the state of Florida, the United States, this study divides the severity of accident injuries into three categories. Random parameter logit models with heterogeneity in both means and variances are developed for the overall ramp area as well as for diverging, merging, and weaving subareas. Choose 22 explanatory variables from the four aspects of driver, vehicle, roadway, and environmental characteristics. The log- likelihood ratio tests are used to evaluate the transferability of different ramp areas model parameters, and the heterogeneity of influencing factors is analyzed through the distribution characteristics of random parameters. At the same time, the direction and degree of influence of significant variables on the severity of accident injury are quantified by using the average elastic coefficient. RESULTS: The results show that there is no transferability of parameters between the overall model and subarea models or among the three ramp subareas, which verifies the necessity of zoning modeling analysis. Different ramp subareas show significantly different means and variances characteristics. "Hit and run: No," "Pre-crash speed: [20, 40)," and "Driver gender: Male" emerge as random parameters in diverging, merging, and weaving areas. Comparative analyses further reveal that some variables exert opposite effects on crash injury severity across different ramp zones, whereas others are significant only within a single subarea. Only a limited number of factors-such as driver condition, driver gender, low pre-crash speed, and vehicle type-consistently influence crash injury severity across all ramp subareas. CONCLUSIONS: The factors affecting the severity of accident injuries in the freeway ramp area have significant spatial heterogeneity. The research results systematically analyzed the significant influencing factors of each subarea, and found that driver behavior, pre-collision speed, road speed limit, lighting conditions all have different effects in different subareas. Regionally differentiated improvement measures and management policies should be formulated for different subareas.
Saudi parents' knowledge, attitude, and behavior regarding child restraint systems during road trips.
Baghdadi LR, Almutairi RM, Almutairi LS … +3 more , Almuslih RA, Alayidh RS, Aldossary GA
Traffic Inj Prev · 2026 Jun · PMID 42246218 · Publisher ↗
OBJECTIVE: The proper use of child restraint systems (CRS) in cars can significantly reduce the risk of injury. This study aimed to assess Saudi parents' knowledge related to CRS use and examined the association between...
OBJECTIVE: The proper use of child restraint systems (CRS) in cars can significantly reduce the risk of injury. This study aimed to assess Saudi parents' knowledge related to CRS use and examined the association between CRS adherence and travel distance. METHODS: A cross-sectional study of Saudi parents with at least one child ageds0-10 years was conducted using a validated electronic questionnaire distributed at public venues throughout Riyadh, using a convenience sampling technique for recruitment. The survey evaluated demographics, knowledge of CRS, and road-safety measures, using a score system and driving patterns related to travel. RESULTS: A total of 385 Saudi parents participated in the study. The average age of participants was 37.69 years ( ± 7.44), 82.6% were women, and 88.8% of participants were aware of the Saudi CRS policy. Booklets and user manuals as sources of CRS information and an increased number of CRS information sources were associated with significantly higher adherence to CRS while traveling in a vehicle ( = 0.040 and = 0.037, respectively). Participants who always used their seat belts also consistently used CRS for their children (70.7%), significantly different from those parents who sometimes/never used a seat belts ( < 0.001). CRS knowledge was significantly higher for parents who always wore seat belts ( = 0.006). Driving speed was a significant factor associated with adherence to CRS and higher speed (> 100 km/h) associated with lower adherence ( = 0.024). CONCLUSION: Increasing parental awareness and providing detailed guidance on the correct installation and usage of CRS in vehicles could help ensure children's safety and well-being.
Behavioral risk profiles in urban pedestrian-vehicle crashes: Insights from latent class and SHAP analysis.
Ergin ME
Traffic Inj Prev · 2026 Jun · PMID 42246216 · Publisher ↗
OBJECTIVES: Traffic crashes remain a major global public health threat, yet most studies focus on crash severity rather than pre-crash factors. This study examines fault attribution as a classification criterion in pedes...
OBJECTIVES: Traffic crashes remain a major global public health threat, yet most studies focus on crash severity rather than pre-crash factors. This study examines fault attribution as a classification criterion in pedestrian-vehicle accidents, separately analyzing accidents where only the driver or only the pedestrian is at fault, and identifying latent behavioral profiles within each stratum using class membership derived from LCA as the primary analytical outcome. METHODS: A hybrid methodology combining Stratified Latent Class Analysis (LCA) and Random Forest (RF)-supported SHapley Additive exPlanations (SHAP) was applied to a dataset of 7,213 vehicle-pedestrian interaction crashes in 2022 and 2023 in Istanbul. Datasets were stratified by fault attribution group (driver fault; pedestrian fault), and LCA was independently applied to each stratum. The optimal number of latent classes was determined using Bayesian Information Criterion (BIC) and Bootstrap Likelihood Ratio Tests (BLRT). RF surrogate models were then trained on LCA-derived class labels using iteratively selected features, and SHAP values were computed to quantify and interpret the contribution of each variable to class membership. RESULTS: Three distinct latent class profiles were identified for each group. In crashes where only pedestrians were attributed fault, traffic control infrastructure-particularly the presence of traffic signals-emerged as the most dominant variable (SHAP ≈ 0.70). In crashes where only drivers were attributed fault, temporal and behavior-based variables played a more decisive role, with seasonal conditions (SHAP ≈ 0.50) and vehicle movement direction (SHAP ≈ 0.45) identified as the primary discriminating parameters. The RF surrogate model achieved classification accuracies of 99.86% and 98.86% for pedestrian and driver strata, respectively. CONCLUSIONS: Fault attribution in urban pedestrian-vehicle crashes is structurally heterogeneous and cannot be adequately captured by global severity models. The results indicate that urban traffic safety management requires not only infrastructure investment but also behavioral intervention strategies tailored to specific user groups. For pedestrian-fault crashes, active signal control measures are critical. For driver-fault crashes, season-specific speed regulations and targeted awareness programs for young drivers are recommended. Additional references can be found in the bibliography in the Appendix.
Speed-dependent temporal development of driver fatigue in extra-long highway tunnels: An on-road study.
Gao XJ, Li YJ, Xu PF … +2 more , Li WY, Sun SY
Traffic Inj Prev · 2026 Jun · PMID 42246215 · Publisher ↗
OBJECTIVE: Prolonged driving in extra-long highway tunnels is characterized by sustained visual monotony and reduced environmental variability, conditions that trigger the insidious onset of driver fatigue. While fatigue...
OBJECTIVE: Prolonged driving in extra-long highway tunnels is characterized by sustained visual monotony and reduced environmental variability, conditions that trigger the insidious onset of driver fatigue. While fatigue in tunnel environments has been widely discussed, limited research has systematically examined how driving speed modulates its temporal development during extended exposure. This study aimed to investigate the speed-dependent characteristics of fatigue development in extra-long tunnels and to identify corresponding temporal and spatial fatigue-prone intervals. METHODS: On-road driving experiments were conducted in an extra-long highway tunnel under three operating speed conditions (60, 80, and 100 km/h). Driver fatigue was assessed using electrocardiography combined with self-report measures. Within the frequency domain of heart rate variability, the low-frequency to high-frequency power ratio (LF/HF) was adopted as the objective indicator of fatigue. Mathematical models were established to quantify the relationship between the LF/HF change rate and driving time under each speed condition, and a fatigue threshold was determined using the 85th-percentile method. These models and the fatigue threshold were then used to identify the onset of fatigue and the time of peak accumulation. Temporal fatigue indicators were further converted into equivalent travel distances to estimate speed-dependent spatial intervals within the tunnel. RESULTS: Driver fatigue in extra-long tunnels manifested a gradual and cumulative progression, characterized by a significant temporal increase in the LF/HF change rate. Driving speed significantly influenced the temporal characteristics of fatigue development. Specifically, the time to reach the peak LF/HF change rate increased with speed, occurring at 176.4 s (60 km/h), 204.2 s (80 km/h), and 256.8 s (100 km/h). No significant differences were observed in LF/HF change rates at fatigue onset across speeds, and a common physiological threshold of 1.42 was determined. Converting temporal fatigue thresholds into travel distances revealed speed-dependent fatigue-prone intervals along the tunnel. CONCLUSIONS: Driving speed modulates the temporal trajectory of fatigue accumulation without changing the physiological boundary of fatigue onset, suggesting that fatigue-related crash risks in extra-long tunnels are intrinsically linked to operating speeds and should be considered in speed-aware safety management strategies.
Psychoactive substance use Among fatally injured traffic crash victims: A cross-sectional study Across four cities in Brazil.
Greve JMD, Bombana HS, de Oliveira RA … +5 more , Pereira MM, Miziara ID, Carvalho HB, Yonamine M, Leyton V
Traffic Inj Prev · 2026 Jun · PMID 42246213 · Publisher ↗
OBJECTIVE: To characterize the prevalence and patterns of psychoactive substance use among fatal traffic crash victims across four Brazilian metropolitan areas and to identify the demographic and temporal factors indepen...
OBJECTIVE: To characterize the prevalence and patterns of psychoactive substance use among fatal traffic crash victims across four Brazilian metropolitan areas and to identify the demographic and temporal factors independently associated with substance positivity. METHODS: Cross-sectional study of 524 fatal traffic crash victims from Recife ( = 272), Curitiba ( = 96), Vitória ( = 93), and Belém ( = 63), March 2022-June 2024. Standardized blood toxicological screening for alcohol, cocaine (benzoylecgonine), benzodiazepines, amphetamines, and cannabis was performed at a single reference laboratory. Multivariable logistic regression identified independent predictors of alcohol and any substance positivity. RESULTS: Overall, 46.0% of victims tested positive for at least one substance. Alcohol was detected in 38.0%, with mean blood alcohol concentration (BAC) of 1.83 g/L (SD = 0.88); no significant difference across cities ( = 0.095). Cocaine or metabolites were detected in 9.9%, with uniform regional distribution ( = 0.392). Vitória showed markedly higher benzodiazepine positivity (9.7% vs. 1.0-4.4%; = 0.019). Cannabis was not detected. Night time (OR = 2.45, 95%CI 1.63-3.67) and weekend crashes (OR = 2.05, 95%CI 1.38-3.04) were the strongest independent predictors of alcohol positivity. After adjustment, Vitória showed significantly higher odds of any substance positivity compared to Recife (OR = 1.73, 95%CI 1.01-2.98, = 0.046). CONCLUSION: Psychoactive substances were involved in nearly half of fatal traffic crashes across four Brazilian cities. Temporal factors-night time and weekend crashes-were the primary independent predictors of substance involvement. Vitória's elevated benzodiazepine positivity persisted after adjustment, suggesting distinct regional prescribing patterns. These findings support temporally targeted enforcement strategies and region-specific pharmaceutical interventions.
Drop the beat, not your guard: Music album releases, smartphone distraction, and traffic fatalities in Chile.
González FAI
Traffic Inj Prev · 2026 May · PMID 42208038 · Publisher ↗
OBJECTIVE: Driver distraction from smartphones is a growing but difficult-to-study contributor to traffic fatalities. We investigated whether major music album releases-events that trigger sharp surges in smartphone-base...
OBJECTIVE: Driver distraction from smartphones is a growing but difficult-to-study contributor to traffic fatalities. We investigated whether major music album releases-events that trigger sharp surges in smartphone-based streaming-are associated with increases in traffic fatalities in Chile, and whether the increase is driven by driver inattention. METHODS: We used 1,145,932 individual-level traffic accident records from Chile's Comisión Nacional de Seguridad de Tránsito (CONASET) for 2009-2023. We conducted a quasi-experimental event study around the release dates of 15 major music albums (2017-2023). We estimated the release-day effect on daily fatalities using OLS with album, day-of-week, and holiday fixed effects, with standard errors clustered at the album level. We exploited two unique features of the Chilean data-police-classified accident causes and detailed accident typologies-to test the distraction mechanism directly. RESULTS: On album release days, daily traffic fatalities increased by 1.80 (95% CI: 0.27-3.34; = .022), a 40.3% increase. Fatalities in crashes officially classified as caused by driver inattention nearly doubled on release days (+90%; = .017); vehicle-vehicle collision fatalities rose by 55% ( = .040), while rollover fatalities-a built-in negative control-showed no change ( = .896). The overall effect was concentrated in daytime, non-alcohol, urban crashes. A permutation-based placebo test ( = .007) and five alternative specifications confirmed robustness. CONCLUSIONS: Music album releases are associated with significant increases in traffic fatalities in Chile, with cause-specific and accident-type evidence directly implicating driver inattention as the mechanism. These findings contribute novel quasi-experimental evidence on smartphone-enabled distraction and road safety from a high-streaming Latin American setting.
Theoretical model and practical application of calibrating cloth method for measuring vehicle speed status.
Zhang W, Zhang Y, Zhao W … +9 more , Zhao X, Yang Y, Yang T, Huang X, Chen J, Ren H, Li D, Zhang M, Miao Q
Traffic Inj Prev · 2026 May · PMID 42208024 · Publisher ↗
OBJECTIVE: To address the issue of calculating vehicle speed from original videos due to unclear reference objects and feature points, this study employs calibration cloth for on-site reexamination and combines image fus...
OBJECTIVE: To address the issue of calculating vehicle speed from original videos due to unclear reference objects and feature points, this study employs calibration cloth for on-site reexamination and combines image fusion technology for video speed identification. METHODS: In the process of constructing an experimental model for vehicle speed measurement using the calibration cloth method, image fusion was performed between the original video footage and the calibration cloth video. The calibration cloth used was a black-and-white checkerboard pattern consisting of 25 rows and 9 columns, with each grid measuring 20 cm × 20 cm. Three vehicle types-sedan, MPV, and SUV-were selected for the experiment. Actual vehicle speeds were calculated across six speed intervals (40 km/h, 50 km/h, 60 km/h, 70 km/h, 80 km/h, and 90 km/h). The speed error rates were analyzed and compared with those obtained from the traditional wheelbase method and the monocular vision velocimetry method based on YOLO+DeepSORT + DepthAnything, thereby verifying the feasibility, accuracy, and advantages of the calibration cloth method. The proposed method was further validated through analysis of real-world accident cases. RESULTS: The proposed method achieved satisfactory image fusion performance, demonstrating good feasibility, stability, and accuracy. Its effectiveness was particularly pronounced within the 40-70 km/h speed range. Compared with the traditional wheelbase method, the calibration cloth method yielded smaller speed estimation errors; compared with the monocular vision velocimetry method, it exhibited superior stability. Furthermore, the method is applicable to speed measurement for various vehicle types and can effectively reflect vehicle motion states. CONCLUSION: This paper proposes a novel vehicle speed measurement method-the calibration cloth method-which is characterized by high accuracy, strong stability, wide applicability, and operational simplicity. The promotion and application of this method will help expand the applicable conditions of video-based speed forensic analysis and improve measurement quality.
Characterizing head acceleration events in Stock Car Auto Racing by head kinematics derived Principal Direction of Force (PDOF).
Smith CD, Mink ZR, Pritchard NS … +5 more , Bullock GS, Harper MG, Urban JE, Patalak JP, Stitzel JD
Traffic Inj Prev · 2026 May · PMID 42208021 · Publisher ↗
OBJECTIVES: Severe brain injury incidence in motorsports has decreased due to driver safety improvements; however, drivers are exposed to non-concussive head acceleration events (HAEs) during active racing maneuvers and...
OBJECTIVES: Severe brain injury incidence in motorsports has decreased due to driver safety improvements; however, drivers are exposed to non-concussive head acceleration events (HAEs) during active racing maneuvers and crashes. The objective of this study was to evaluate whether Principal Direction of Force (PDOF) influences head impact exposure in National Association for Stock Car Auto Racing (NASCAR) Cup Series drivers. METHODS: Head kinematic data was collected from 20 NASCAR Cup Series drivers during 41 races in the 2024 season using instrumented mouthpiece sensors. Sensors collected linear and rotational head kinematics (1600 Hz, 4 g threshold) and were characterized as crash (i.e., a driver's vehicle interacted with an external object, resulting in a caution) or race (i.e., normal, green flag racing) events and by track type. HAEs with multiple impacts were manually segmented into individual impacts. Impacts were assigned to one of eight PDOF directions characterizing frontal, rear, side, and oblique impacts. Resultant linear acceleration was calculated in the transverse plane with pulse boundaries determined using thresholds (5 g if peak >10 g; 2.5 g if peak ≤10 g). PDOF was calculated from changes in linear velocity in the transverse plane using these boundaries. Peak linear and rotational accelerations and velocities (PLA, PRA, PRV), change in linear velocity (ΔLV), and Diffuse Axonal Multi-Axis General Evaluation (DAMAGE) were calculated per impact. Mixed effects models evaluated associations between PDOF and head kinematics, adjusting for fixed effects of impact classification, track type, and random effects of driver, followed by pairwise comparisons among PDOF directions. RESULTS: 3,156 HAEs yielded 5,198 impacts; 720 impacts not exceeding 4 g in the transverse plane were excluded, leaving 4,478 impacts (Race:3,818 (85%); Crash 660 (15%)) for analysis. Right side impacts were most common ( = 2,476, 55.3%) with left frontal oblique impacts least frequent ( = 60, 1.3%). Predicted means from pairwise comparisons revealed directional patterns. For PLA, frontal (8.28 g [7.59,9.04]), right side (8.61 g [8.29,8.95]), and rear (9.09 g [8.56,9.65]) impacts produced greater magnitudes than left side (6.72 g [6.43,7.02]) and oblique directions. PRA showed similar patterns, with right side (331 rad/s [304, 360]) and rear (347 rad/s [314, 383]) having the greatest magnitudes. PRV was greatest in frontal (4.44 rad/s [4.00, 4.94]) and frontal oblique impacts (4.29 rad/s [3.86,4.76] right; 4.19 rad/s [3.70,4.75] left). DAMAGE was greatest in right frontal oblique (0.036 [0.032, 0.041]) impacts. Only DAMAGE saw an oblique direction produce the highest magnitudes. CONCLUSIONS: PDOF was associated with head impact frequency and magnitude in NASCAR Cup Series drivers. Right side impacts were both most frequent and among the highest magnitude, while frontal oblique impacts showed greater effects on estimated brain tissue deformation. This is the first study to characterize directionality using PDOF derived directly from head kinematics. Findings provide a framework for evidence-based, targeted safety interventions in NASCAR, other motorsports, and contact sports.
Reclined postures when changing both seat back and seat pan angles.
Grébonval C, Beillas P, Wang X
Traffic Inj Prev · 2026 May · PMID 42208019 · Publisher ↗
OBJECTIVE: The study aims to predict body as a function of seat pan (SPA) and seat back (SBA) angles for safety and comfort evaluation in the context of development of highly automated vehicles. METHODS: Data were collec...
OBJECTIVE: The study aims to predict body as a function of seat pan (SPA) and seat back (SBA) angles for safety and comfort evaluation in the context of development of highly automated vehicles. METHODS: Data were collected from thirteen participants testing eleven SBA and SPA combinations with SBA and SPA respectively varying from 20° to 60° (relative to the vertical) and from 14° to 40° (relative to the horizontal) using a reconfigurable experimental seat. Participants were instructed to adjust seat parameters for a relaxed seating position except those imposed by a test condition. Using the position of reflective markers captured by a motion capture system, manually palpated anatomical landmarks on the pelvis, and ischial tuberosities by pressure measurement, a global optimization method was applied to reconstruct body postures with help of a personalized kinematic model including the head, spine, pelvis and two lower limbs. RESULTS: Regression analysis showed that both SBA and SPA affect not only the orientation of body segments as expected, but also relative angles between segments (except cervical flexion). Different pelvic and spinal postures should be used even when laterally rotating the whole seat altogether with a constant seat back to seat pan angle. CONCLUSIONS: The proposed regression equations could be used for positioning postmortem human surrogates (PMHS) for physical tests or human body models (HBM) for safety or ergonomics digital simulations.
Hybrid modeling of unobserved heterogeneity in Ohio freeway crash severity using RWIS data: A Random Forest-Correlated Mixed Logit approach.
Bukuru E, Balyagati PF, Sando T … +2 more , Kidando E, Owusu-Danquah J
Traffic Inj Prev · 2026 May · PMID 42201311 · Publisher ↗
OBJECTIVE: Weather-related crashes continue to contribute disproportionately to severe injuries and fatalities on high-speed freeways despite advances in Intelligent Transportation Systems (ITS). This study evaluates the...
OBJECTIVE: Weather-related crashes continue to contribute disproportionately to severe injuries and fatalities on high-speed freeways despite advances in Intelligent Transportation Systems (ITS). This study evaluates the safety impacts of real-time weather conditions on freeway crash injury severity across Ohio using integrated Road Weather Information System (RWIS), crash, and roadway data. METHODS: Hourly RWIS observations, crash records, and roadway characteristics from 2019-2023 were integrated, covering approximately 452,292 freeway crashes. A hybrid analytical framework combining Random Forest variable selection and Correlated Mixed Logit with Heterogeneity in Means (CMXL-HM) modeling was applied to capture unobserved heterogeneity in driver responses. An hourly exposure-based Severity Index (SI) was developed to normalize crash frequencies across weather states and quantify relative severity risk under varying environmental conditions. RESULTS: Severity Index results indicated that freezing temperatures and moderate wind speeds exhibited the highest relative severity risk. The CMXL-HM model significantly outperformed baseline Multinomial Logit and Mixed Logit models based on likelihood ratio tests (χ = 21,754.69, < 0.001) and goodness-of-fit measures. Freezing temperatures (≤32 °F) and wind speed (5-10 mph) were identified as random parameters with statistically significant heterogeneity, while other parameters exhibited more consistent effects. Several adverse weather conditions exhibited protective effects after controlling traffic and roadway factors, suggesting risk-compensating driver behavior. Marginal effects showed that higher traffic volumes and rear-end collisions increased crash severity, while morning periods and the spring season reduced it. A strong and statistically significant positive association (ρ = 0.937) between freezing temperatures and travel speed indicates compounded high-risk conditions. CONCLUSIONS: Findings indicate that real-time weather conditions exert heterogeneous and nonlinear effects on freeway crash severity. The proposed hourly exposure-normalized framework provides actionable insights for identifying high-risk weather thresholds and supports evidence-based deployment of RWIS infrastructure, weather-responsive Variable Speed Limits, and targeted traveler information systems to enhance freeway safety management.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe