Next Article in Journal
Global Performance Index for Integrated Management System: GPI-IMS
Next Article in Special Issue
Risky Driver Recognition with Class Imbalance Data and Automated Machine Learning Framework
Previous Article in Journal
A Generalized Structural Equation Model Approach to Long Working Hours and Near-Misses among Healthcare Professionals in Japan
Previous Article in Special Issue
Modelling the Relationship between the Nature of Work Factors and Driving Performance Mediating by Role of Fatigue
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 18
Issue 13
10.3390/ijerph18137155
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Assessment of the Influence of Technology-Based Distracted Driving on Drivers’ Infractions and Their Subsequent Impact on Traffic Accidents Severity

by
Susana García-Herrero
1,*,
Juan Diego Febres
2,
Wafa Boulagouas
1,
José Manuel Gutiérrez
3 and
Miguel Ángel Mariscal Saldaña
1
1
Escuela Politécnica Superior, Universidad de Burgos, 09006 Burgos, Spain
2
Department of Chemistry and Exact Sciences, Universidad Técnica Particular de Loja, 110107 Loja, Ecuador
3
Consejo Superior de Investigaciones Científicas, 39005 Santander, Spain
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2021, 18(13), 7155; https://0-doi-org.brum.beds.ac.uk/10.3390/ijerph18137155
Submission received: 30 April 2021 / Revised: 27 June 2021 / Accepted: 29 June 2021 / Published: 4 July 2021
(This article belongs to the Special Issue Driving Behaviors and Road Safety)
Browse Figure
Versions Notes

Abstract

:
Multitasking while driving negatively affects driving performance and threatens people’s lives every day. Moreover, technology-based distractions are among the top driving distractions that are proven to divert the driver’s attention away from the road and compromise their safety. This study employs recent data on road traffic accidents that occurred in Spain and uses a machine-learning algorithm to analyze, in the first place, the influence of technology-based distracted driving on drivers’ infractions considering the gender and age of the drivers and the zone and the type of vehicle. It assesses, in the second place, the impact of drivers’ infractions on the severity of traffic accidents. Findings show that (i) technology-based distractions are likely to increase the probability of committing aberrant infractions and speed infractions; (ii) technology-based distracted young drivers are more likely to speed and commit aberrant infractions; (iii) distracted motorcycles and squad riders are found more likely to speed; (iv) the probability of committing infractions by distracted drivers increases on streets and highways; and, finally, (v) drivers’ infractions lead to serious injuries.

1. Introduction

The road transportation system presents a high-risk system that threatens people’s lives every day [1]. The severity of traffic accidents raises many social direct and indirect problems, physical and mental health disorders, economic expenses, and massive damage to the surroundings and properties [2]. The World Health Organization estimates that approximately 1.35 million people die in road traffic accidents each year (on average, 3700 people lose their lives per day), and 20 to 50 million more people suffer non-fatal injuries, which often lead to long-term disabilities [3]. In this regard, studies on the features of road safety concluded that traffic accidents occur mainly due to human factors, road infrastructure, environmental aspects, and their interactions [4,5]. Nevertheless, a growing body of research suggests that human factors, e.g., speeding, drink-driving, distracted driving, have the strongest influence and are responsible for 80–90% of road traffic accidents [6,7,8,9]. Indeed, past research reported that distracted driving contributes to over half of inattention traffic accidents [10]. This is supported by an explosion of studies. For instance, in the USA, the National Highway Safety Administration estimated that distracted driving is responsible for approximately 10% of all fatal traffic accidents [11]. Similarly, in Spain, the General Traffic Department, reported that distracted driving contributed to approximately 28% of police-reported fatal traffic accidents [12].
Placing greater importance on the fact that secondary tasks that compete for the driver’s attention and potentially degrade their perception and their ability to interpret the information they receive continually from a changing roadway [13,14], in the recent past, many researchers and traffic transportation experts have deeply studied aspects of distracted driving and concluded that a majority of distracted driving is related to new technologies. Furthermore, a review of the current literature on distracted driving found a specific focus on mobile phone-related distractions. For instance, a survey study reported that 42% of drivers confirmed that they answer their mobile phones when driving, and 56% admitted to continuing to drive while completing the conversation [15]. An observational study of 6578 drivers on randomly selected urban roads in Spain found that 20% of the observed drivers engaged in secondary tasks including talking on handheld mobile phones (i.e., 1.3%) [16].
A simulator study investigating the relationship between performing a secondary task (e.g., mobile phone use) and driving performance reported that engaging in a secondary task influences longitudinal and lateral control of the vehicle and leads to higher speeds [17]. Moreover, an examination of the chance of drivers colliding increases twelvefold when they handled their mobile phones [18]. Furthermore, it has been found that the effects of the use of mobile phones on drivers’ reactions are worse than driving under a 0.08% blood alcohol level [19].
Another experimental study analyzed the effects of the use of mobile phones on the reaction time of drivers in dangerous situations (pedestrian crossing and road crossing by parked vehicles, in particular) found that the use of mobile phones leads up to 204% increments in reaction times, proving that distractions decrease the driving performance [20].
However, an estimation of the contribution of distracted driving in traffic accidents causality and its impact on other unsafe driving behaviors is complex. Indeed, there are the following three major challenges [21,22]: (i) obtaining reliable data about pre-accident conditions is difficult, (ii) there is a lack of systemic reporting, and (iii) there are inconsistencies in the definitions, classifications, and approaches used. Dealing with these challenges, many scholars have reviewed traffic accident assessment tools and advanced techniques and algorithms to increase the efficiency and effectiveness of road safety protective measures [23,24,25]. Recently, many researchers have moved toward correlating the analysis methods with prediction techniques to model interactions between risk factors and predict potential impacts on causalities, frequencies, and the severity of traffic accidents. Such a combination considers several parameters to analyze the current conditions that are, therefore, assessed using the mean of the prediction models that contribute to mitigate the magnitude of traffic accidents and enhance the transportation system and safety strategies [26]. Among the emergent techniques, there are the Grey System Theory and Markov Model [2], Data Mining Techniques [27], Structural Time Series [28], Logistic Regression Analysis [29], and Bayesian Networks [30].
Building on these attempts, this paper focuses particularly on analyzing the influence of technology-based distracted driving on drivers’ infractions and assesses their subsequent impact on the severity of traffic accidents employing recent data on road traffic accidents in Spain.
The present study is designed to provide relevant pieces of evidence on the relationship between technology-based distracted driving and other infractions. The assessment is therefore extended to evaluate the relationships between the infractions of the distracted drivers and the severity of traffic accidents and investigate the impact of a set of factors grouped into demographics, type of vehicle, and zone.
An assessment of the influence of technology-based distractions on drivers’ infractions allows the accident risk to be estimated. In other words, it allows for, first, an appreciation of the proportion by which the probability of committing aberrant infractions and speeding would be expected to increase, provided that the driver is distracted. Second, it captures the infractions that lead to serious traffic accidents resulting in fatalities, while incorporating relevant parameters.
The remainder of this paper is arranged as follows: Section 2 reviews distracted driving; Section 3 sums up the data and methodology of the study; Section 4 provides the results of the study; Section 5 discusses the results and puts forward main findings, limitations, and future research guideline, while Section 6 concludes the paper.

2. Background

Distracted driving has emerged as a major phenomenon that compromises traffic safety. It adversely impacts driving performance, increases reaction time, and reduces control over the vehicle; thus, it accounts for 25% of severe motor vehicle accidents, leading to significant morbidity and mortality [31]. In terms of definition, distracted driving refers to the inattention of drivers and their focus on other competing activities while operating a motor vehicle, for instance, talking, smoking, texting, putting on make-up, reading, eating, using mobile phones, etc. [32].
In light of the past research, distracted driving could be either intentional and voluntary, which occurs when drivers divert their attention from the driving tasks, or involuntary due to a failure to ignore non-related stimuli that motivates drivers to become distracted [33,34,35].
Distractors have been grouped into the following four main categories [36,37]: (i) visual, which implies taking the eyes off the road; (ii) auditory, which prevents making the best use of hearing; (iii) manual, which considers taking the hands off the wheel; and, finally, (iv) cognitive, when losing concentration on driving.
Moreover, distractions have been classified into [38,39,40] (i) in-vehicle distractions, such as using mobile phones, interacting with an entertainment system, an iPod, radio, DVD player, operating navigation system, etc., and (ii) on-road distractions, such as roadside advertisements, crash scenes, digital billboards, etc.
Although other potential distractions are still of interest, technology-based distractions have attracted the focus of researchers who thoroughly investigated the risks associated with the use of technology devices (e.g., smartphones, wearable devices, portable devices, and in-vehicle information systems) while driving and used many approaches for risk estimation, for instance [41], surveys, simulators, phone records, on-road testing, and naturalistic studies.
A study analyzing the contribution of distracted driving to traffic accidents in the United States reported that 16% of motor vehicle crashes in which people were killed and 20% of those that caused injuries were a result of distracted driving [13]. Furthermore, a synthesis of previous studies on mobile phone use and its effects on traffic safety has identified mobile phones as leading sources of distraction that reduce driving performance as they decrease the reaction time of the drivers [17,42,43,44]. After investigating the driving performance of distracted drivers, many studies have concluded that distracted drivers are more likely to engage in unsafe driving behaviors and several errors and violations increase, for instance, speeding violations, right-of-way violations, and failure to stop at stop signs and red lights [45,46,47]. For a better understanding of the characteristics of distracted drivers and the frequency of being engaged in distractions, many researchers have investigated the demographics of distracted drivers. Indeed, in terms of gender, a study conducted in the UK reported that no differences have been found for many types of distractions (e.g., mobile phone conversations) [48]. Furthermore, the results of this study considered the age of the driver as a crucial predictor for most of the studied distractions (including technology-based distractions), concluding that older drivers are less likely to be distracted, unlike younger drivers. Similarly, a roadside observational survey in Melbourne [49] has found that gender does not influence distracted driving, in contrast to age, where it has been suggested that young or middle-aged drivers are the predominant group in distracted driving.
Second, despite the increasing body of research on the effects of distracted driving on traffic accidents, a large part of the literature relies mostly on survey, observational, and simulator studies, which suffer a set of limitations. First, survey studies, most of the time, target one particular group of drivers (e.g., young, male, female, etc.) or analyze distracted driving by accumulating all the groups together. Furthermore, the data collected in such survey and/or interview studies are likely to be biased [50]. Even though observational studies are typically conducted at particular places, allowing sufficient time for the observer to capture the behaviors of the drivers, they, in turn, limit the data to one location along the roadway [51]. Finally, driving simulator experiments are safe assessment procedures and provide a more realistic environment; however, traffic situations are unpredictable and uncontrollable and, therefore, simulators have limited fidelity [52].
The present study is designed to adequately remove, first, the bias of underreporting and after that deploys a machine learning technique to adequately investigate the influence of distracted driving on drivers’ infractions considering the drivers’ characteristics, the type of vehicle, and the zone, and assess the impact on the severity of traffic accidents.

3. Materials and Methods

3.1. Data Description

For this study, a dataset has been prepared using official data of traffic accidents that occurred in Spain in a period of four years (2016–2019) provided by the Spanish National Transportation Department. The data have been gathered by the Civil Guard General Directorate or local police officers and include information related to the traffic accident, for instance, the date, location, type of vehicles involved in the traffic accidents, demographics of the drivers, number of fatalities, trip purpose, violations, and errors, etc.
The current sample of this analysis includes 410,974 traffic accidents involving 666,504 drivers. Details of the study sample are given in Table 1.
The dataset contains information about the severity of traffic accidents presented in (Table 2). This information is aggregated into the following two groups: (i) Fatal accidents (FA) resulting in serious injuries (SI) or fatalities to the drivers involved in serious accidents (55545), and (ii) Minor accidents (MA) causing no injuries (NI) or resulting in slight injuries to the drivers involved in these minor accidents.
The collected information from the dataset about technology-based distractions and drivers’ infractions is given in Table 3.

3.2. Bias Identification

The dataset used to conduct this study was gathered by the Civil Guard General Directorate and local police officers who register information on traffic accidents, which is always incomplete. Indeed, many researchers have confirmed that as data on traffic accidents involving distracted drivers are collected from traffic accident reports, the real influence of this later on driving behaviors goes underestimated [53,54]. Furthermore, distractions are hard to prove using statistics from the police [55], because, in general, police officers only look for distractions when the consequences of traffic accidents are serious and, therefore, many distracted driving cases may not be recorded. Thus, scientific studies lead to unreliable conclusions.
To address the reporting biases, a methodology proposed in recent research [46] has been adopted. It is based on the introduction of a “dummy variable” into the model to isolate homogeneous subsamples and generate valid model and unbiased parameter estimations. Accordingly, the frequencies of “technology-based distractions” have been thoroughly analyzed. Particularly, differences between the percentage of drivers involved in severe accidents knowing the states (i.e., being or not distracted) and the percentage of drivers involved in severe accidents having unknown states (i.e., unspecified) are computed (10.86% versus 7.40%, respectively) (Table 4). The differences are found to be significant; therefore, the variable “technology-based distractions” is biased, and a dummy variable is introduced to differentiate homogeneous cases. In the rest of the paper, the sensitivity analysis results will only show the probabilities of drivers’ infractions in traffic accident severity in a case of the state, “presence or absence of technology-based distractions” of dummy variable technology-based distractions. In this way, the subsamples containing unknown cases about the technology-based distractions are eliminated.

3.3. Bayesian Networks

In the present study, Bayesian Networks have been deployed to model the influence of distracted driving on the severity of traffic accidents and unsafe driving behaviors. Over the last few decades, the Bayesian approach has been extensively applied to traffic safety studies, for instance, an assessment of road safety [56,57], an assessment of the influence of seatbelt use on the severity of traffic accidents [58], an estimation of committing an infraction due to mobile use [59], music distraction among young drivers [60], a prediction of traffic accidents [61], etc.
The Bayesian Networks model is a graphical inference methodology capable of predicting the future behavior of a particular variable based on past experience learned from the historical data source. Furthermore, they consist mainly of the following:
  • The qualitative aspect of the Bayesian Networks given by a directed acyclic network generally denoted as DAG (V, E), consisting of nodes (V) representing the variables related with directed edges (E), denoting the dependencies between the variables;
  • The quantitative aspect consists of the conditional probability of each node, where every node has parents and has a conditional probability table expressing the dependencies of the father nodes. Therefore, the joint probability distribution is expressed as follows:
    P ( X 1 , … ,   X n ) = ∏ i = 1 n P ( X 1 | X i − 1 , … , X n )
    P ( X ) = ∏ i = 1 n P ( X i | p a i )  
    where i = 1 , 2 , … ,   n , X = { X 1 ,   X 2 ,   … , X n } is a set of variables of the Bayesian Network, P is a set of local probability distributions associated with each variable, X i refers to the variable node and, p a i is the father node of X i .
Many types of software could be used to build the Bayesian network, among them we cite the following [62]: BayesBuilder, JavaBayes, and Bayes Net Toolbox, which were used in the present study.

3.4. Network Validation

To measure the performance of the obtained Bayesian Network, a ten-fold cross-validation has been conducted. K-fold cross-validation is a powerful means of testing the success rate, accuracy, and robustness of models used for classification [63]. It consists of randomly partitioning the dataset into ten subsets of equal size; then each subset, in turn, is used to validate the model fitted on the remaining k-1 subsets. The evaluation of the obtained models is performed using the Area Under the ROC (Receiver Operating Characteristic) Curve—named AUC—that plots the following two parameters: True Positive Rate and False Positive Rate that yield the performance of a classification model. The standard scores range from 0 (opposite and wrong prediction) to 0.5 (random prediction) and 1 (perfect prediction).

4. Results

4.1. Validation of Bayesian Network

The results of the performance evaluation of the learned Bayesian Network (Figure 1) are given in Table 5.
The AUC score metrics obtained a range from 0.62 to 0.99 for the variables, which confirms the high robustness of the Bayesian Network.
The directed acyclic graph of Figure 1 specifies the joint distribution and represents the dependences/independences between the study variables. Indeed, in the lower right part of the graph, the largest number of study variables such as the zone, gender, vehicle type, and age can be observed in a grouped way. These variables have direct relationships with the infractions’ variables before relating, directly or indirectly, to the severity of the accident. At the top of the graph, the infractions’ variables are directly related to the technology-based distraction variable. Furthermore, aberrant and speed infractions are related to the dummy variable, which is directly related to the severity of the traffic accidents.

4.2. Sensitivity Analysis

4.2.1. Assessment of the Influence of Technology-Based Distractions on Drivers’ Infractions

Estimation of the “a priori” probabilities of drivers’ infractions considering the technology-based distractions is given by the sensitivity analysis results in Table 6.
The results in Table 6 show that the probability of committing aberrant infractions increases significantly (from 23.57 to 66.75%) given the fact that the drivers are distracted. Interestingly, these results show that the probability of not committing aberrant infractions increased notably (from 33.25 to 76.43%) when the drivers are not distracted.
Similarly, the sensitivity analysis results in Table 6 show that drivers are more likely to speed when they are distracted. Indeed, the probability increases from 7.78 to 14.45%. Moreover, the probability of respecting the speed limit increases from 85.55% to 92.22% when drivers are not distracted.
The results also show that technology-based distractions are more likely to affect aberrant infractions than speed infractions.

4.2.2. Assessment of the Influence of Technology-Based Distractions on Drivers’ Infractions Considering the Age and Gender of the Drivers

The influence of distracted driving on drivers’ infractions considering the effect of demographics is given by the sensitivity analysis results in Table 7.
According to these results, the probability of not committing aberrant infractions, considering the fact that the drivers are not distracted, increases for drivers younger than 60 years old. Similarly, it has been found that the probability of not speeding, considering the fact that the drivers are not distracted, increases for drivers older than 40 years old and more significantly in the case of elderly drivers (from 92.22 to 95.39%).
With regard to the gender, the results in Table 8 show that the probability of not speeding, considering the fact that the drivers are not distracted, increases notably in the case of females (from 92.22 to 94.08%).
In terms of aberrant infractions, the sensitivity analysis results show that the probabilities increase given the fact that the drivers are distracted (from 66.75 to 68%) in the case of younger drivers (<25 years old). Similarly, it has been found that the probability of speeding for distracted drivers increases, interestingly, in the case of younger drivers (<25 years old) (from 14.45 to 27.14%) and decreases significantly in the case of older drivers (from 14.45 to 8.28%).

4.2.3. Assessment of the Influence of Technology-Based Distractions on Drivers’ Infractions Considering the Zone and Type of the Vehicle

The results of the estimation of the influence of technology-based distractions on drivers’ infractions considering the zone and type of the vehicle are given in Table 8. The results show that the probability that drivers do not commit aberrant infractions increases in the case of motorcycles, quads, and quadricycles (from 76.43 to 86.58%) and other types of vehicles (from 76.43 to 85.48%), considering the fact that the drivers are not distracted.
In contrast, the results show that the probability of committing aberrant infractions increases more significantly in the case of distracted drivers of cars, vans, and all-terrain vehicles (from 66.75 to 68.91%).
With regard to speed, similarly, the results of the sensitivity analysis show that the probability of respecting the speed limits increases significantly in the case of non-distracted motorcycle riders and drivers of other vehicles (from 92.22 to 93.46%, and from 92.22 to 95.67%, respectively). Moreover, the probability of speeding increases, logically, in the case of distracted motorcycle riders.
The results of the effect of the zone on the behaviors of distracted drivers confirm that distracted drivers commit aberrant infractions on highways and streets (the probabilities increase from 66.75 to 79.80% and from 66.75 to 81.11%, respectively). Nevertheless, distracted drivers are more likely to not respect the speed limit on highways (the probability increases significantly from 14.45 to 39.59%).

4.2.4. Assessment of the Influence of Drivers’ Infractions on Traffic Accident Severity

The severity of traffic accidents due to aberrant infractions or speeding is not always reported in the case of minor traffic accidents. In other words, in such accidents, the police officers are not used to thoroughly fill in the traffic accident report and, most of the time, they do not provide details on whether the drivers involved in minor accidents had committed aberrant infractions or speed infractions.
The influence of drivers’ infractions on the severity of traffic accidents is given by the sensitivity analysis results in Table 9.
The results in Table 9 show that the probability that the severity of injuries is serious, resulting in fatalities, increases, given the fact that the drivers committed aberrant infractions (from 9.38 to 9.69%) and more significantly in the case of speeding (the probability increases from 9.03 to 17.89%).

5. Discussions

Traffic accidents have become a major health public issue and preservation of the lives of drivers, passengers and users is among the main concerns of communities around the world. The literature revealed the fact that numerous factors contribute to and influence the occurrence of these accidents and the severity of injuries [64,65,66]. Although the continuous efforts of governments and numerous traffic safety policies being issued to control traffic accidents, the rates of disabilities, fatalities, and injuries continue to increase dramatically. Therefore, it has been extensively recommended to analyze the different risk factors influencing traffic accident severity to yield more patterns and polish knowledge to effectively and efficiently prevent road accidents and ameliorate traffic safety.
In this study, we focused, particularly, on the assessment of the influence of technology-based distractions on the performance of drivers behind the wheel. The contribution of this research is twofold. First, it deployed machine learning technique algorithms that, taking advantage of advancements in information technology, allow traffic accidents to be predicted and detect the role of different risk factors in traffic accident scenarios. Second, it overviewed the relationships between the technology-based distractions and the infractions of drivers with full consideration of the impact of gender, age, zone, and the type of vehicle.
Given the crucial aspect of traffic safety, i.e., the prediction of the most important risk factors impacting the occurrence of traffic accidents and influencing the severity of the outcomes, in the present study, a predictive model has been developed to assess the influence of technology-based distractions on the traffic accidents’ severity and unsafe driving behaviors considering the effect of a set of selected risk factors using Bayesian Networks methodology. Indeed, the obtained Bayesian model presented the knowledge in the form of joint probabilities distributions of the study variables, which is vital to make effective and efficient decisions that avoid the drivers’ infractions resulting from technology-based distractions and reduce the severity of injuries to the drivers and passengers.
To conduct this investigation, a recent database over four years (from 2016–2019) on traffic accidents that occurred in Spain was used. According to the sensitivity analyses, the findings indicated that technology-based distracted driving has a significant effect on both the aberrant infractions and speeding. Indeed, the study found that the presence of distractions notably increases the probability of committing aberrant infractions and speeding. This is in line with the literature [67,68] that overviewed the characteristics of distracted driving, which has agreed that new technologies can absorb drivers’ attention and reduce their abilities to judge driving demands and disrespect driving safety requirements that lead, most of the time, to aberrant infractions and unsafe behaviors.
With regard to the risk factors related to the demographics of the drivers involved in traffic accidents, vehicle, and zone, the findings of the present study reported increased probabilities as regards the infractions of distracted young drivers (under 25 years old). These findings coincide with many conclusions of existing studies [69,70,71], which had reported similar observations on risky driving behaviors and violations of young drivers who are more susceptible to be involved in fatal crashes due to distraction activities (e.g., mobile phone use, radio, DVD, etc.). In this context, young drivers are inexperienced and easily distracted by interactions with music devices, texting and conversations on mobile phones resulting in slower reaction times [72].
The sensitivity analysis results of the present study also showed that there are no significant differences in the probabilities of the impact of the gender of drivers on the driving behaviors of distracted drivers.
Furthermore, it has been found that speeding will lead to severe traffic accidents, leading to serious injuries and/or fatalities. These results are tightly associated with the findings of many previous researchers [73,74] that confirmed that speed increases the chance of traffic accidents that result in greater severity.
With regard to the influence of the zone and type of the vehicle factors, the sensitivity analysis showed that distracted drivers are more likely to speed on highways, whereas on streets, distracted drivers are more susceptible to commit aberrant infractions. Moreover, it has been found that the probability of motorcycle riders speeding increases because they are distracted, while car drivers are more likely to commit aberrant infractions. Similar findings reported that mortalities on highways and outside the urban center are mainly due to speed-related crashes [75,76]. Furthermore, an investigation into drivers’ violations found that speeding infractions represent 80% of registered traffic violations on highways [77].
For future research, it is recommended to extend the current study by considering other risk factors. Such consideration would give wider context and more shreds of evidence on the influence of distracted driving that would help improve and enhance traffic safety more effectively and efficiently.

6. Conclusions

Distracted driving is a growing threat to road safety and accounts for a significant number of serious injuries and deaths in traffic accidents. In this paper, the distracting effects of various technologies on driving performance have been assessed using Bayesian Networks. Unlike other studies, the assessment was then extended to analyze the influence of the driver’s infractions on the severity of traffic accidents. The design of this study considered the driver’s characteristics, the type of vehicle, and zone.
The results of this study documented a strong link between technology-based distracted driving and aberrant and speed infractions. Moreover, the findings showed that these infractions have a direct impact on the severity of traffic accidents. Furthermore, this study specifically found that young drivers are more likely to be distracted.
The deployment of Bayesian Networks yielded a representative graphical structure of the relationships among the technology-based distractions, drivers’ infractions, and traffic accidents’ severity. It contributed to overcoming the observed limitations of most research that has used regression models. Thus, machine learning techniques proved to be more suitable for prediction models in traffic safety problems.

Author Contributions

Conceptualization, S.G.-H., J.D.F. and W.B.; methodology, J.M.G.; software, J.D.F.; validation, J.D.F. and S.G.-H.; formal analysis, J.D.F. and W.B.; investigation, W.B. and J.D.F.; resources, S.G.-H.; data curation, J.D.F.; writing—original draft preparation, W.B., J.D.F. and S.G.-H.; writing—review and editing, W.B., J.D.F., S.G.-H., M.Á.M.S. and J.M.G.; supervision, S.G.-H. and J.M.G.; project administration, S.G.-H.; funding acquisition, S.G.-H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the project “Modelización mediante técnicas de machine learning de la influencia de las distracciones del conductor en la seguridad vial”, grant number BU300P18, and was funded by FEDER (Fondo Europeo de Desarrollo Regional—Junta de Castilla y León).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Acknowledgments

We would like to thank the Dirección General de Tráfico from España (DGT) for providing the data for this study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Le, K.G.; Liu, P.; Lin, L.-T. Determining the Road Traffic Accident Hotspots Using GIS-Based Temporal-Spatial Statistical Analytic Techniques in Hanoi, Vietnam. Geo-Spat. Inf. Sci. 2020, 23, 153–164. [Google Scholar] [CrossRef] [Green Version]
  2. Jin, X.; Zheng, J.; Geng, X. Prediction of Road Traffic Accidents Based on Grey System Theory and Grey Markov Model. Int. J. Saf. Secur. Eng. 2020, 10, 263–268. [Google Scholar] [CrossRef]
  3. WHO. Available online: https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries (accessed on 18 March 2021).
  4. Soehodho, S. Public Transportation Development and Traffic Accident Prevention in Indonesia. IATSS Res. 2017, 40, 76–80. [Google Scholar] [CrossRef] [Green Version]
  5. Cai, Q. Cause Analysis of Traffic Accidents on Urban Roads Based on an Improved Association Rule Mining Algorithm. IEEE Access 2020, 8, 75607–75615. [Google Scholar] [CrossRef]
  6. Petridou, E.; Moustaki, M. Human Factors in the Causation of Road Traffic Crashes. Eur. J. Epidemiol. 2000, 16, 819–826. [Google Scholar] [CrossRef] [PubMed]
  7. Ledesma, R.D.; Montes, S.A.; Poó, F.M.; López-Ramón, M.F. Individual Differences in Driver Inattention: The Attention-Related Driving Errors Scale. Traffic Inj. Prev. 2010, 11, 142–150. [Google Scholar] [CrossRef]
  8. Alavi, S.S.; Mohammadi, M.R.; Souri, H.; Mohammadi Kalhori, S.; Jannatifard, F.; Sepahbodi, G. Personality, Driving Behavior and Mental Disorders Factors as Predictors of Road Traffic Accidents Based on Logistic Regression. Iran. J. Med. Sci. 2017, 42, 24–31. [Google Scholar]
  9. Boulagouas, W.; García-Herrero, S.; Chaib, R.; Febres, J.D.; Mariscal, M.Á.; Djebabra, M. An Investigation into Unsafe Behaviors and Traffic Accidents Involving Unlicensed Drivers: A Perspective for Alignment Measurement. Int. J. Environ. Res. Public Health 2020, 17, 6743. [Google Scholar] [CrossRef]
  10. Stutts, J.; Reinfurt, D.; Staplin, L.; Rodgman, E. The Role of Driver Distraction in Traffic Crashes. AAA Found. Traffic Saf. 2001, 202. [Google Scholar]
  11. Xiong, H.; Bao, S.; Sayer, J.R. Factors Affecting Drivers’ Cell Phone Use Behavior: Implications from a Naturalistic Study. Transp. Res. Rec. J. Transp. Res. Board 2014, 2434, 72–79. [Google Scholar] [CrossRef]
  12. Dirección General de Tráfico. Las Principales Cifras de la Siniestralidad Vial España 2019; Dirección General de Tráfico: Madrid, Spain, 2019; ISBN 2445-0219.
  13. Overton, T.L.; Rives, T.E.; Hecht, C.; Shafi, S.; Gandhi, R.R. Distracted Driving: Prevalence, Problems, and Prevention. Int. J. Inj. Control Saf. Promot. 2015, 22, 187–192. [Google Scholar] [CrossRef] [PubMed]
  14. Ngueutsa, R.; Kouabenan, D.R. Fatalistic Beliefs, Risk Perception and Traffic Safe Behaviors. Eur. Rev. Appl. Psychol. 2017, 67, 307–316. [Google Scholar] [CrossRef]
  15. Schroeder, P.; Wilbur, M.; Pena, R. National Survey on Distracted Driving Attitudes and Behaviors—2015; National Highway Traffic Safety Administration: Washington, DC, USA, 2018; p. 194.
  16. Prat, F.; Planes, M.; Gras, M.E.; Sullman, M.J.M. An Observational Study of Driving Distractions on Urban Roads in Spain. Accid. Anal. Prev. 2015, 74, 8–16. [Google Scholar] [CrossRef]
  17. Oviedo-Trespalacios, O.; Haque, M.M.; King, M.; Washington, S. Understanding the Impacts of Mobile Phone Distraction on Driving Performance: A Systematic Review. Transp. Res. Part C Emerg. Technol. 2016, 72, 360–380. [Google Scholar] [CrossRef] [Green Version]
  18. Dingus, T.A.; Guo, F.; Lee, S.; Antin, J.F.; Perez, M.; Buchanan-King, M.; Hankey, J. Driver Crash Risk Factors and Prevalence Evaluation Using Naturalistic Driving Data. Proc. Natl. Acad. Sci. USA 2016, 113, 2636–2641. [Google Scholar] [CrossRef] [Green Version]
  19. Strayer, D.L.; Drews, F.A.; Crouch, D.J. A Comparison of the Cell Phone Driver and the Drunk Driver. Hum. Factors J. Hum. Factors Ergon. Soc. 2006, 48, 381–391. [Google Scholar] [CrossRef] [PubMed]
  20. Choudhary, P.; Velaga, N.R. Modelling Driver Distraction Effects due to Mobile Phone Use on Reaction Time. Transp. Res. Part C Emerg. Technol. 2017, 77, 351–365. [Google Scholar] [CrossRef]
  21. Beanland, V.; Fitzharris, M.; Young, K.L.; Lenné, M.G. Driver Inattention and Driver Distraction in Serious Casualty Crashes: Data from the Australian National Crash In-Depth Study. Accid. Anal. Prev. 2013, 54, 99–107. [Google Scholar] [CrossRef] [PubMed]
  22. Wundersitz, L. Driver Distraction and Inattention in Fatal and Injury Crashes: Findings from in-Depth Road Crash Data. Traffic Inj. Prev. 2019, 20, 696–701. [Google Scholar] [CrossRef]
  23. Yannis, G.; Dragomanovits, A.; Laiou, A.; La Torre, F.; Domenichini, L.; Richter, T.; Ruhl, S.; Graham, D.; Karathodorou, N. Road Traffic Accident Prediction Modelling: A Literature Review. Proc. Inst. Civ. Eng. Transp. 2017, 1–10. [Google Scholar] [CrossRef]
  24. Lord, D.; Mannering, F. The Statistical Analysis of Crash-Frequency Data: A Review and Assessment of Methodological Alternatives. Transp. Res. Part A Policy Pract. 2010, 44, 291–305. [Google Scholar] [CrossRef] [Green Version]
  25. Gutierrez-Osorio, C.; Pedraza, C. Modern Data Sources and Techniques for Analysis and Forecast of Road Accidents: A Review. J. Traffic Transp. Eng. 2020, 7, 432–446. [Google Scholar] [CrossRef]
  26. Akgüngör, A.P.; Doğan, E. An Artificial Intelligent Approach to Traffic Accident Estimation: Model Development and Application. Transport 2009, 24, 135–142. [Google Scholar] [CrossRef]
  27. Gupta, M.; Solanki, V.K.; Singh, V.K. Analysis of Datamining Technique for Traffic Accident Severity Problem: A Review. In Proceedings of the Second International Conference on Research in Intelligent and Computing in Engineering, Uttrakhand, India, 9 June 2017; pp. 197–199. [Google Scholar]
  28. Yousefzadeh-Chabok, S.; Ranjbar-Taklimie, F.; Malekpouri, R.; Razzaghi, A. A Time Series Model for Assessing the Trend and Forecasting the Road Traffic Accident Mortality. Arch. Trauma Res. 2016, 5. [Google Scholar] [CrossRef] [Green Version]
  29. Hu, F.; Lv, D.; Zhu, J.; Fang, J. Related Risk Factors for Injury Severity of E-Bike and Bicycle Crashes in Hefei. Traffic Inj. Prev. 2014, 15, 319–323. [Google Scholar] [CrossRef] [PubMed]
  30. Hongguo, X.; Huiyong, Z.; Fang, Z. Bayesian Network-Based Road Traffic Accident Causality Analysis. In Proceedings of the 2010 WASE International Conference on Information Engineering, Beidai, China, 14–15 August 2010; pp. 413–417. [Google Scholar]
  31. Lee, J.D. Dynamics of Driver Distraction: The Process of Engaging and Disengaging. Ann. Adv. Automot. Med. 2014, 58, 24–32. [Google Scholar] [PubMed]
  32. Donmez, B.; Boyle, L.N.; Lee, J.D. The Impact of Distraction Mitigation Strategies on Driving Performance. Hum. Factors J. Hum. Factors Ergon. Soc. 2006, 48, 785–804. [Google Scholar] [CrossRef]
  33. Feng, J.; Marulanda, S.; Donmez, B. Susceptibility to Driver Distraction Questionnaire: Development and Relation to Relevant Self-Reported Measures. Transp. Res. Rec. J. Transp. Res. Board 2014, 2434, 26–34. [Google Scholar] [CrossRef] [Green Version]
  34. Marulanda, S.; Chen, W.; Donmez, B. Capturing Voluntary, Involuntary, and Habitual Components of Driver Distraction in a Self-Reported Questionnaire. In Proceedings of the 2015 Driving Assessment Conference, Salt Lake City, UT, USA, 1 January 2015. [Google Scholar]
  35. Chen, H.-Y.W.; Hoekstra-Atwood, L.; Donmez, B. Voluntary- and Involuntary-Distraction Engagement: An Exploratory Study of Individual Differences. Hum. Factors J. Hum. Factors Ergon. Soc. 2018, 60, 575–588. [Google Scholar] [CrossRef]
  36. Sundfør, H.B.; Sagberg, F.; Høye, A. Inattention and Distraction in Fatal Road Crashes—Results from In-Depth Crash Investigations in Norway. Accid. Anal. Prev. 2019, 125, 152–157. [Google Scholar] [CrossRef] [PubMed]
  37. Le, A.S.; Suzuki, T.; Aoki, H. Evaluating Driver Cognitive Distraction by Eye Tracking: From Simulator to Driving. Transp. Res. Interdiscip. Perspect. 2020, 4, 100087. [Google Scholar] [CrossRef]
  38. Ito, H.; Atsumi, B.; Uno, H.; Akamatsu, M. Visual Distraction While Driving. IATSS Res. 2001, 25, 20–28. [Google Scholar] [CrossRef] [Green Version]
  39. Hudák, M.; Madleňák, R. The Research of Driver Distraction by Visual Smog on Selected Road Stretch in Slovakia. Procedia Eng. 2017, 178, 472–479. [Google Scholar] [CrossRef]
  40. Shaaban, K.; Gaweesh, S.; Ahmed, M.M. Investigating In-Vehicle Distracting Activities and Crash Risks for Young Drivers Using Structural Equation Modeling. PLoS ONE 2020, 15, e0235325. [Google Scholar] [CrossRef]
  41. Hill, L.; Rybar, J.; Styer, T.; Fram, E.; Merchant, G.; Eastman, A. Prevalence of and Attitudes About Distracted Driving in College Students. Traffic Inj. Prev. 2015, 16, 362–367. [Google Scholar] [CrossRef] [PubMed]
  42. Stavrinos, D.; Jones, J.L.; Garner, A.A.; Griffin, R.; Franklin, C.A.; Ball, D.; Welburn, S.C.; Ball, K.K.; Sisiopiku, V.P.; Fine, P.R. Impact of Distracted Driving on Safety and Traffic Flow. Accid. Anal. Prev. 2013, 61, 63–70. [Google Scholar] [CrossRef] [Green Version]
  43. Llerena, L.E.; Aronow, K.V.; Macleod, J.; Bard, M.; Salzman, S.; Greene, W.; Haider, A.; Schupper, A. An Evidence-Based Review: Distracted Driver. J. Trauma Acute Care Surg. 2015, 78, 147–152. [Google Scholar] [CrossRef]
  44. Zatezalo, N.; Erdogan, M.; Green, R.S. Road Traffic Injuries and Fatalities among Drivers Distracted by Mobile Devices. J. Emerg. Trauma Shock 2018, 11, 175–182. [Google Scholar] [CrossRef]
  45. Huisingh, C.; Griffin, R.; McGwin, G. The Prevalence of Distraction among Passenger Vehicle Drivers: A Roadside Observational Approach. Traffic Inj. Prev. 2015, 16, 140–146. [Google Scholar] [CrossRef] [Green Version]
  46. Useche, S.A.; Alonso, F.; Montoro, L.; Esteban, C. Distraction of Cyclists: How Does It Influence Their Risky Behaviors and Traffic Crashes? PeerJ 2018, 6, e5616. [Google Scholar] [CrossRef]
  47. García-Herrero, S.; Gutiérrez, J.M.; Herrera, S.; Azimian, A.; Mariscal, M.A. Sensitivity Analysis of Driver’s Behavior and Psychophysical Conditions. Saf. Sci. 2020, 125, 104586. [Google Scholar] [CrossRef]
  48. Sullman, M.J.M.; Prat, F.; Tasci, D.K. A Roadside Study of Observable Driver Distractions. Traffic Inj. Prev. 2015, 16, 552–557. [Google Scholar] [CrossRef]
  49. Young, K.L.; Rudin-Brown, C.M.; Lenné, M.G. Look Who’s Talking! A Roadside Survey of Drivers’ Cell Phone Use. Traffic Inj. Prev. 2010, 11, 555–560. [Google Scholar] [CrossRef]
  50. Qi, Y.; Vennu, R.; Pokhrel, R. Distracted Driving: A Literature Review; Illinois Center for Transportation: Rantoul, IL, USA, 2020. [Google Scholar]
  51. Wenners, K.E.; Knodler, M.A. Examination of Limitations Associated with Observing Driver Distraction. Transp. Res. Rec. J. Transp. Res. Board 2014, 2434, 135–139. [Google Scholar] [CrossRef]
  52. Karthaus, M.; Wascher, E.; Getzmann, S. Distraction in the Driving Simulator: An Event-Related Potential (ERP) Study with Young, Middle-Aged, and Older Drivers. Safety 2021, 7, 36. [Google Scholar] [CrossRef]
  53. Farmer, C.M.; Braitman, K.A.; Lund, A.K. Cell Phone Use While Driving and Attributable Crash Risk. Traffic Inj. Prev. 2010, 11, 466–470. [Google Scholar] [CrossRef]
  54. Bucsuházy, K.; Matuchová, E.; Zůvala, R.; Moravcová, P.; Kostíková, M.; Mikulec, R. Human Factors Contributing to the Road Traffic Accident Occurrence. Transp. Res. Procedia 2020, 45, 555–561. [Google Scholar] [CrossRef]
  55. Wegman, F. The Future of Road Safety: A Worldwide Perspective. IATSS Res. 2017, 40, 66–71. [Google Scholar] [CrossRef] [Green Version]
  56. Grande, Z.; Castillo, E.; Mora, E.; Lo, H.K. Highway and Road Probabilistic Safety Assessment Based on Bayesian Network Models: Highway and Road Probabilistic Safety Assessment. Comput. Aided Civ. Infrastruct. Eng. 2017, 32, 379–396. [Google Scholar] [CrossRef]
  57. Mora, E.; Grande, Z.; Castillo, E. Bayesian Networks Probabilistic Safety Analysis of Highways and Roads. In Computer Aided Systems Theory—EUROCAST 2017; Moreno-Díaz, R., Pichler, F., Quesada-Arencibia, A., Eds.; Springer: Cham, Switzerland, 2018; Volume 10672, pp. 470–476. ISBN 978-3-319-74726-2. [Google Scholar]
  58. Febres, J.D.; García-Herrero, S.; Herrera, S.; Gutiérrez, J.M.; López-García, J.R.; Mariscal, M.A. Influence of Seat-Belt Use on the Severity of Injury in Traffic Accidents. Eur. Transp. Res. Rev. 2020, 12. [Google Scholar] [CrossRef] [Green Version]
  59. Catalina Ortega, C.A.; García-Herrero, S.; Azimian, A.; García, A.; Mariscal, M.A. Estimating the Probability of Committing Traffic Infraction due to Mobile Use: Driving Simulator Study; Research Publishing Services: Chennai, India, 2020; pp. 2585–2591. [Google Scholar]
  60. Catalina, C.A.; García-Herrero, S.; Cabrerizo, E.; Herrera, S.; García-Pineda, S.; Mohamadi, F.; Mariscal, M.A. Music Distraction among Young Drivers: Analysis by Gender and Experience. J. Adv. Transp. 2020, 2020, 1–12. [Google Scholar] [CrossRef]
  61. Alizadeh, S.S.; Mortazavi, S.; Sepehri, M.M. Prediction of Vehicle Traffic Accidents Using Bayesian Networks. Sci. J. Pure Appl. Sci. 2014, 3. [Google Scholar] [CrossRef]
  62. Zou, X.; Yue, W.L. A Bayesian Network Approach to Causation Analysis of Road Accidents Using Netica. J. Adv. Transp. 2017, 2017, 1–18. [Google Scholar] [CrossRef] [Green Version]
  63. Marcot, B.G.; Hanea, A.M. What Is an Optimal Value of k in K-Fold Cross-Validation in Discrete Bayesian Network Analysis? Comput. Stat. 2020. [Google Scholar] [CrossRef]
  64. Sanyang, E.; Peek-Asa, C.; Bass, P.; Young, T.L.; Daffeh, B.; Fuortes, L.J. Risk Factors for Road Traffic Injuries among Different Road Users in the Gambia. J. Environ. Public Health 2017, 2017, 1–9. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  65. Xie, S.; Ji, X.; Yang, W.; Fang, R.; Hao, J. Exploring Risk Factors with Crash Severity on China Two-Lane Rural Roads Using a Random-Parameter Ordered Probit Model. J. Adv. Transp. 2020, 2020, 1–14. [Google Scholar] [CrossRef]
  66. Hosseinian, S.M.; Najafi Moghaddam Gilani, V.; Tahmasbi Amoli, H.; Nikookar, M.; Orouei, A. Presentation of Analytical Methods for Better Decision Making about the Most Important Factor Influencing Rural Accidents. Math. Probl. Eng. 2021, 2021, 1–16. [Google Scholar] [CrossRef]
  67. Foss, R.D.; Goodwin, A.H. Distracted Driver Behaviors and Distracting Conditions among Adolescent Drivers: Findings from a Naturalistic Driving Study. J. Adolesc. Health 2014, 54, S50–S60. [Google Scholar] [CrossRef] [Green Version]
  68. Eid, H.O.; Abu-Zidan, F.M. Distraction-Related Road Traffic Collisions. Afr. Health Sci. 2017, 17, 491–499. [Google Scholar] [CrossRef] [Green Version]
  69. Híjar, M.; Carrillo, C.; Flores, M.; Anaya, R.; Lopez, V. Risk Factors in Highway Traffic Accidents: A Case Control Study. Accid. Anal. Prev. 2000, 32, 703–709. [Google Scholar] [CrossRef]
  70. Ali, E.K.; El-Badawy, S.M.; Shawaly, E.-S.A. Young Drivers Behavior and Its Influence on Traffic Accidents. J. Traffic Logist. Eng. 2014, 2, 45–51. [Google Scholar] [CrossRef]
  71. Tanveer, M.; Kashmiri, F.A.; Naeem, H.; Yan, H.; Qi, X.; Rizvi, S.M.A.; Wang, T.; Lu, H. An Assessment of Age and Gender Characteristics of Mixed Traffic with Autonomous and Manual Vehicles: A Cellular Automata Approach. Sustainability 2020, 12, 2922. [Google Scholar] [CrossRef] [Green Version]
  72. Buckley, L.; Chapman, R.L.; Sheehan, M. Young Driver Distraction: State of the Evidence and Directions for Behavior Change Programs. J. Adolesc. Health 2014, 54, S16–S21. [Google Scholar] [CrossRef] [Green Version]
  73. Ratanavaraha, V.; Suangka, S. Impacts of Accident Severity Factors and Loss Values of Crashes on Expressways in Thailand. IATSS Res. 2014, 37, 130–136. [Google Scholar] [CrossRef] [Green Version]
  74. Lee, J.; Chae, J.; Yoon, T.; Yang, H. Traffic Accident Severity Analysis with Rain-Related Factors Using Structural Equation Modeling—A Case Study of Seoul City. Accid. Anal. Prev. 2018, 112, 1–10. [Google Scholar] [CrossRef]
  75. Cooper, P.J. The Relationship between Speeding Behaviour (as Measured by Violation Convictions) and Crash Involvement. J. Saf. Res. 1997, 28, 83–95. [Google Scholar] [CrossRef]
  76. Valent, F.; Schiava, F.; Savonitto, C.; Gallo, T.; Brusaferro, S.; Barbone, F. Risk Factors for Fatal Road Traffic Accidents in Udine, Italy. Accid. Anal. Prev. 2002, 34, 71–84. [Google Scholar] [CrossRef]
  77. Ahmed, M.; Sahnoon, I.; Al-Zaidy, A. Predicting Speed-Related Traffic Violations on Rural Highways. In Proceedings of the 2nd World Congress on Civil, Structural, and Environmental Engineering (CSEE’17), Barcelona, Spain, 2–4 April 2017. [Google Scholar]
Ijerph 18 07155 g001 550
Figure 1. Obtained directed acyclic graph corresponding to the study variables.
Figure 1. Obtained directed acyclic graph corresponding to the study variables.
Ijerph 18 07155 g001
Table
Table 1. Frequencies of the study sample.
Table 1. Frequencies of the study sample.
VariablesNumber of Cases
Traffic AccidentVehicle Type2016201720182019Total
Car, van, all-terrain vehicles120,831120,261119,755118,491479,338
Motorcycles, quads, quadricycles35,22236,05136,31937,467145,059
Heavy vehicles890689019176887335,856
Other vehicles922759103935316251
Total165,881165,972166,289168,362666,504
Drivers
Demographics		Age
Y < 25 21,98321,35020,70721,43285,472
25 ≤ Y ≤ 40 63,18861,47661,19160,538246,393
40 ≤ Y ≤ 60 59,85761,66063,69665,369250,582
Y > 60 17,39918,10218,33319,01072,844
Unspecified 345433842362201311,213
Total 165,881165,972166,289168,362666,504
Gender
Male 119,878120,447120,920122,407483,652
Female 445,4444,22544,77245,407178,948
Unspecified 145913005975483904
Total 165,881165,972166,289168,362666,504
Table
Table 2. Traffic accident severity distribution.
Table 2. Traffic accident severity distribution.
Traffic
Accident
Severity		Severity LevelNumber of CasesTotal
2016201720182019
M/NI151,446151,595152,757155,161610,959
SI/F14,43514,37713,53213,20155,545
Total165,881165,972166,289168,362666,504
SI/F: Serious Injuries and/or Fatalities M/NI: Minor and/or No Injuries.
Table
Table 3. Frequencies of technology-based distractions and drivers’ infractions.
Table 3. Frequencies of technology-based distractions and drivers’ infractions.
InfractionsDrivers’ InfractionsNumber of Cases
2016201720182019Total
Aberrant infractions
No infractions54,40552,13152,05462,566221,156
Aberrant infractions34,55835,62336,26040,083146,524
Unspecified76,91878,21877,97565,713298,824
Total165,881165,972166,289168,362666,504
Speed infractions
No speed infractions70,57369,45167,25279,677286,953
Speed infractions895781548395811733,623
Unspecified86,35188,36790,64280,568345,928
Total165,881165,972166,289168,362666,504
Distracted DrivingTechnology-based
distractions
No distractions41,76641,79041,94442,634168,134
Technology-based distractions8811029102411144048
No technology-based distractions or unspecified123,234123,153123,321124,614494,322
Total165,881165,972166,289168,362666,504
Table
Table 4. Dummy Variable Frequencies.
Table 4. Dummy Variable Frequencies.
Dummy Variable Technology-Based DistractionsStatesNumber of CasesTotalPercentageSI/F
2016201720182019
Presence or absence of technology-based distractions42,64742,81942,96843,748172,18225.83%10.86%
Unknown123,234123,153123,321124,614494,32274.17%7.40%
Total165,881165,972166,289168,362666,504/
Table
Table 5. Validation results of the 10-fold cross-validation.
Table 5. Validation results of the 10-fold cross-validation.
VariablesAccident SeverityAberrant InfractionsSpeed InfractionsTechnology-Based Distraction
StatesSI/FM/NINoYesNoYesNoYes
AUC scores0.620.620.880.770.900.770.990.93
Table
Table 6. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions.
Table 6. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions.
Technology-Based DistractionsAberrant InfractionsSpeed Infractions
States No Yes No Yes
No76.43%23.57%92.22%7.78%
Yes33.25%66.75%85.55%14.45%
Table
Table 7. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions considering the influence of demographics.
Table 7. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions considering the influence of demographics.
DemographicsTechnology-Based DistractionsAberrant InfractionsSpeed Infractions
AgeStatesNoYesNoYes
Y < 25No76.65%23.35%85.66%14.34%
Yes32.00%68.00%72.86%27.14%
25 ≤ Y ≤ 40No76.50%23.50%91.39%8.61%
Yes33.18%66.82%84.10%15.90%
40 ≤ Y ≤ 60No76.61%23.39%93.96%6.04%
Yes33.82%66.18%88.94%11.06%
Y > 60No75.54%24.46%95.39%4.61%
Yes33.66%66.34%91.72%8.28%
GenderTechnology-Based DistractionsAberrant InfractionsSpeed Infractions
StatesNoYesNoYes
MaleNo76.38%23.62%91.45%8.55%
Yes33.21%66.79%84.25%15.75%
FemaleNo76.56%23.44%94.08%5.92%
Yes33.45%66.55%88.91%11.09%
Table
Table 8. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions considering the influence of the zone and vehicle type.
Table 8. Sensitivity analysis for the influence of technology-based distractions on drivers’ infractions considering the influence of the zone and vehicle type.
VariablesTechnology-Based DistractionsAberrant InfractionsSpeed Infraction
Vehicle TypeStatesNoYesNoYes
Car/Van/All TerrainNo73.73%26.27%92.02%7.98%
Yes31.09%68.91%85.80%14.20%
Motorcycles/quads/quadricyclesNo86.58%13.42%93.46%6.54%
Yes43.83%56.17%84.56%15.44%
Heavy vehiclesNo78.28%21.72%90.44%9.56%
Yes39.70%60.30%84.73%15.27%
Other vehiclesNo85.48%14.52%95.67%4.33%
Yes34.90%65.10%85.37%14.63%
ZoneTechnology-Based DistractionsAberrant InfractionsSpeed Infraction
StatesNoYesNoYes
RoadNo76.19%23.81%89.03%10.97%
Yes41.06%58.94%85.64%14.36%
Street or similarNo76.74%23.26%96.97%3.03%
Yes18.89%81.11%85.50%14.50%
HighwayNo83.82%16.18%94.19%5.81%
Yes20.20%79.80%60.41%39.59%
Table
Table 9. Sensitivity analysis for the influence of the drivers’ infractions on the severity of traffic accidents.
Table 9. Sensitivity analysis for the influence of the drivers’ infractions on the severity of traffic accidents.
Drivers’ InfractionsSeverity of Traffic Accidents
Aberrant infractionsM/NISI/F
No90.62%9.38%
Yes90.31%9.69%
Speed infractions
No90.97%9.03%
Yes82.11%17.89%
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

García-Herrero, S.; Febres, J.D.; Boulagouas, W.; Gutiérrez, J.M.; Mariscal Saldaña, M.Á. Assessment of the Influence of Technology-Based Distracted Driving on Drivers’ Infractions and Their Subsequent Impact on Traffic Accidents Severity. Int. J. Environ. Res. Public Health 2021, 18, 7155. https://0-doi-org.brum.beds.ac.uk/10.3390/ijerph18137155

AMA Style

García-Herrero S, Febres JD, Boulagouas W, Gutiérrez JM, Mariscal Saldaña MÁ. Assessment of the Influence of Technology-Based Distracted Driving on Drivers’ Infractions and Their Subsequent Impact on Traffic Accidents Severity. International Journal of Environmental Research and Public Health. 2021; 18(13):7155. https://0-doi-org.brum.beds.ac.uk/10.3390/ijerph18137155

Chicago/Turabian Style

García-Herrero, Susana, Juan Diego Febres, Wafa Boulagouas, José Manuel Gutiérrez, and Miguel Ángel Mariscal Saldaña. 2021. "Assessment of the Influence of Technology-Based Distracted Driving on Drivers’ Infractions and Their Subsequent Impact on Traffic Accidents Severity" International Journal of Environmental Research and Public Health 18, no. 13: 7155. https://0-doi-org.brum.beds.ac.uk/10.3390/ijerph18137155

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop