Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Effect of Physical Environment Risk Factors on Vehicle Collisions Severity Involving Child-Pedestrians in Malaysia:

The Effect of Physical Environment Risk Factors on Vehicle Collisions Severity Involving... This study is aimed at investigating the association between child-pedestrian severity levels of collisions and physical environmental variables. The outcome of this study could be applied to road safety intervention for improving engineering modifications related to children pedestrians. The retrospective analysis was carried out using 6-year data from Royal Malaysia Police records from the years 2009 to 2014. Multinomial logit modeling (MNL) was applied. The results demonstrated that the injury severity of the collisions is related to road geometry, road surface material, road surface condition, traffic system, road marking, traffic control type, lighting condition, speed limit, time of collision, type of location, and land use characteristics. Specifically, fatal injury collisions are significantly increased by t/y intersection; concrete and earth-road surfaces; two-way traffic and dual carriageways; posted speed limits of 70 to 90 km/h; time of collision: 0 to 0659 hours (early morning) and 0700 to 0959 hours (morning); lighting conditions, including dark without street light, and dark with street light; and control type involving police. Meanwhile, cross intersections’ posted speed limits of 80 km/h to 90 km/h and the time of collision from 0 to 0659 hours (early morning) and 1900 to 2459 hours (night), significantly increased serious injury collisions. Notably, the findings revealed the importance of more in-depth studies on physical environmental features that relate to child-pedestrians’ severity level of collisions. This is essential for improvements to physical environmental designs by policymakers. Thus, policymakers and stakeholders can utilize the findings to further improve the physical environment through structure and design. Keywords child-pedestrian, road injuries, collisions, severity children under the age of 10 are unable to engage with the Introduction traffic environment owing to their weak sensory and cogni- Child-pedestrians are considered to be one of the most vul- tive skills. Tiwari (2020) also highlighted that conventional nerable road users (VRUs) in non-motorized traffic groups pedestrian instruction is ineffective for young children, par- (Dissanayake et al., 2009). Child-pedestrians are highly ticularly those under the age of 10, whereas earlier research vulnerable to traffic injury due to their underdeveloped has shown a link between cognitive abilities and pedestrian physical, cognitive, developmental, behavioral, and sensory age (Barton et al., 2012; Pitcairn & Edlmann, 2000; Sandels, functions (Barton et al., 2012; Oxley et al., 1997; Schieber & 1970; Whitebread & Neilson, 2000). For example, Barton Thompson, 1996; Sminkey, 2008; Vinje, 1981). In general, child-pedestrians are still developing, resulting in a high Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia incidence of accidents and severe injuries. For instance, their Universiti Teknologi Mara, Shah Alam, Selangor, Malaysia diminutive size impairs their ability to see and be noticed by Malaysian Institute of Road Safety Research (MIROS), Kajang, Selangor, other road users (Oxley et al., 2012; Schieber & Thompson, Malaysia 1996; Solagberu et al., 2014). Consequently, this increases Corresponding Author: the risks of severe and fatal injury (Rothman et al., 2010; Yao Muhamad Nazri Borhan, Department of Civil Engineering/Sustainable et al., 2006). Urban Transport Research Centre (SUTRA), Faculty of Engineering & Growing research indicates that children’s cognitive Built Environment, Universiti Kebangsaan Malaysia, development has an effect on child-pedestrians’ safety. In a Bangi, Selangor 43600, Malaysia. Email: mnazri_borhan@ukm.edu.my study done by Sandels (1970), the author concluded that Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open et al. (2012) investigated the relationship between children’s road conditions, the number of traffic lanes, the volume of age, gender, the two elements of cognitive functioning (visual traffic, play areas, and buildings, are examples of a man- search and efficiency), and child-pedestrians’ route choices. made environment (Peek-Asa & Zwerling, 2003; Wazana The study discovered that males, particularly younger chil- et al., 1997). Meanwhile, weather, extreme temperatures, and dren, choose riskier pedestrian routes. Nonetheless, many lighting are part of the natural environment. studies have shown that other development skills, including Numerous risk variables in the physical environment perceptual and motor abilities that overlap with cognitive have been linked to the severity of accidents between abilities, affect the safety of child-pedestrian interactions child-pedestrians and vehicles. These include road and (Pitcairn & Edlmann, 2000; Schwebel et al., 2012). traffic characteristics, weather and lighting conditions, the However, according to the World Health Organization passage of time, the play area, and the neighborhood (WHO) (Peden, 2004), child-pedestrians of all ages are at (Agran et al., 1996; Davis & Rice, 1994; Dissanayake risk of being involved in a road traffic collision. WHO also et al., 2009; Grechkin et al., 2013; Hobday & Knight, highlighted that 38% of children under 19 years old are 2010; Koopmans et al., 2015; Leden et al., 2006; Steinbach injured or killed as pedestrians on the roads each year et al., 2010). For example, Koopmans et al. (2015) exam- (Toroyan, 2015). In Malaysia, the statistics of child-pedes- ined pedestrian collisions in Illinois, US, using a multi- trian fatalities in Malaysia are particularly worrying due to a variate logistic regression model to estimate injury severity. high rate of fatal road collisions (Mohamed et al., 2011). The The findings showed that risk variables related to physical authors highlighted that 12% of children aged 1 to 18 years settings, such as lighting condition, darkness, midblock old were involved in fatal road collisions from 2007 to 2009. position, traffic management, and road quality, were linked This figure is higher than that of high-income countries, with injury severity in children. Additionally, the authors which only have 5% to 10% of fatalities (Sminkey, 2008). In investigated the interplay of two factors: intersection and particular, a recent study by Darus et al. (2018) found a total traffic control. The results indicate that being located in the of 2,243 child-pedestrian (aged 0–18 years) casualties were middle of a block without traffic control has a substantial reported between 2009 and 2012, accounting for about effect on the severity of injuries. 27.9% of the total number of pedestrian casualties. As a Meanwhile, in a case control study done by Stevenson result, this scenario implies that child-pedestrians are among et al. (1995), Australia, the volume of traffic, in combination the most susceptible road users to traffic injury. with the proportion of vehicles exceeding the speed limit, and As stated by the World Health Organization (WHO) the presence of footpaths, significantly increased the likeli- (Sminkey, 2008), a distinguishing aspect of the physical hood of injuries to child-pedestrians. Moreover, in a review environment, particularly the road environment, is the unique study done by Wazana et al. (1997), the volume of traffic (13 contributing variables associated with children. By and large, times), the speed limit (6.0 times), the predominant type of the road environment is constructed by adults for the benefit dwelling (up to 5.5 times), the absence of a play area (5.3 of adults. As a result, this scenario has increased the danger times), the location on the road (4.2 times), the protection of of harm for children. the play area (3.5 times), the proportion of curbside parking (3.4 times), the street mean vehicle speed (3.3 times), the shared driveway (3.2 times), the type of road (2.9 times), and Physical Environment Risk Factors the time of day (up to 2.7 times) were all found to be signifi- The physical environment refers to the elements of the physi- cant in increasing the probability of child-pedestrian injury. cal surroundings that contribute to the occurrence of poten- Thus, the primary objective of this research is to ascertain tial injury or produce events that cause injury (Haddon, the impact of physical and traffic environment characteris- 1980; Runyan, 2003). The physical environment can be clas- tics on the severity of accidents involving child-pedestrians sified as natural or man-made (Peek-Asa & Zwerling, 2003). aged 18 years or younger in Malaysia. The models are built In the Haddon Matrix model developed by Willam Haddon specifically utilizing multinomial logistic analysis (MNL), Jr., the physical environment is one of the important risk fac- which include critical information about the physical envi- tors within the conceptual framework in all types of injury ronment. The models produced may be used to inform inter- control (Haddon, 1980; Peek-Asa & Zwerling, 2003; vention and environmental change efforts aimed at reducing Runyan, 2003). While, Peek-Asa and Zwerling also pointed injury severity. out that one of the best injury prevention strategies is by changing the physical environment. Generally, by under- Materials and Methods standing the association between the natural/man-made envi- ronment and injury risk, it may suggest significant avenues Data Source and Method of Analysis for injury intervention (Peek-Asa & Zwerling, 2003). In relation to road injuries, the physical characteristics of The dataset was extracted from the Royal Malaysia Police the roadway, such as road conditions, the number of traffic (RMP) by the Malaysian Institute of Road Safety Research lanes, the road’s location, the speed limit, and the volume of (MIROS). The dataset consists of physical environmental Darus et al. 3 Table 1. Terms Defined by the Royal Malaysian Police (RMP) for Road Collisions (Royal Malaysia Police, 2014). Term Definition Pedestrian Any person who is not in a vehicle but occupies a portion of the road; including road construction workers, a person pushing a broken-down vehicle, etc. Fatal road accident A road accident in which one or more people were killed within 30 days of the event’s date. Serious injury road accident A road accident in which at least one person sustained serious injury but none was killed. Slightly/minor injury road accident A road accident in which one or more people were injured, but none were killed or seriously injured. Non-injury road accident A road accident in which no person was killed or injured. characteristics which provide information on the roadway International Comparison of Road Accident characteristics, weather and traffic conditions, and the colli- Database sion time and location. An international comparison database may lead to an increase A total of 2,518 vehicle collisions involving child-pedes- in our understanding of road safety and subsequent counter- trians (events based) were analyzed after excluding about measures in different countries. For this purpose, Table 2 was 4.6% of the total records due to discrepancies such as over- adapted from Montella et al. (2013) and Casado-Sanz et al. lapping report numbers, incomplete records, etc. The analysis (2019) to describe the international comparison between focused on collisions involving child-pedestrians (≤18 years) national databases of selected countries (US, New Zealand, aged 18 years and below for a 6-year period from January Australia, Spain databases including the requirement of EU 2009 to December 2014 in 14 states of Malaysia. Directive). Twenty eight variables were considered for the In this study, the multinomial logit model (MNL) and the comparison. It is noted that in Malaysia, police reports do not associated relative risk ratio (RRR) are obtained and esti- link to the hospital data. Also, specific information such as mated using the Statistical Package for the Social Sciences road segment gradient, average daily traffic (ADT), curve (SPSS) version 23. The Small-Hsiao test of IIA assumption radius and length are not included in the database. is estimated using the Stata 17 software package. Theory of Multinomial Logit Model Malaysia Accident Database The severity levels of the collision were adopted as the Data source dependent variable. These severity levels were defined and The source of data was recorded by the Royal Malaysia recorded by the Royal Malaysian Police (RMP) as shown in Police (RMP) for all types of road accidents. RMP started Table 1. A total of three levels of severity are considered, using the revised accident reporting forms known as Form which are slight/minor injury collision, serious injury colli- POL 27 nationwide on 1st January 1992 (Saidon & Baguley, sion and fatal injury collision. In this study, a multinomial 1994). Overall, the form includes 91 variables, as reported logit model was used to increase flexibility in the model by Ahmed et al. (2020). Basically, the form “POL 27” is specification (Rifaat et al., 2011; Tay et al., 2011). The model divided into eight sections which include reference details of was developed as described in previous studies (Çelik & the report/time of occurrence, carriageway details, environ- Oktay, 2014; Rifaat et al., 2011; Shankar et al., 1996; Tay ment, location, details of the vehicle, details of the driver, et al., 2011; Ulfarsson & Mannering, 2004). details of passengers and pedestrians, and also sketch dia- The probability of a child-pedestrian collision nth with a grams of the accident and location. Then, the data will be severity level i, is as follows: keyed in manually into the Computerized Accident Recording System (CARS) (Hashim & Rahim, 2009). ′ ′ (1) PP =≥ UU ,, ∀i Ii ≠ I () ni ni ni Definitions. In Malaysia, severity of collision is divided where I denotes a set of possible mutually exclusive severity into four levels; fatal, serious, slight, and non-injury. There levels of collisions. are no pedestrian-vehicle accidents recorded as non-injury The U can be expressed as follows: collisions in this data set. As a result, only three severity ni levels are evaluated. The current research makes use of the Ux =+ βε (2) ni in ni definitions set out in Table 1. In contrast to worldwide sta- where β denotes a vector of coefficients, while x denotes a tistics, the majority of nations utilized the same definition vector of explanatory variables, and ε denotes an unobserv- of a fatal traffic accident as Malaysia, namely (dying within ni able random error. ε is assumed to have a generalized 30 days after a crash occurs). However, there are no estab- ni extreme value distribution. From the assumption, the MNL lished definitions of (severe) injuries, and they may vary can be expressed as follows: across nations (OECD, 2010). 4 Table 2. International Comparison in the Road Accident Database. Casado-Sanz et al. (2019). a b Variable EU Directive US MMUCC Australia New Zealand Spain Malaysia Crash location Precise as possible Road name, GPS coordinates Road name, reference Road name, GPS Road name, km Road name, km, location point, distance, direction coordinates GPS coordinates Crash narrative No No Yes Yes Yes Yes Crash sketch No No Yes, access restricted Yes Yes Yes Crash type Yes Recorded in the traffic units Yes Yes Yes Yes section Collision type Yes 8 descriptors Yes Yes 33 descriptors Yes Contributing No Environmental circumstances Yes Yes Yes Yes circumstances Weather conditions Yes 10 descriptors Yes 5 descriptors 9 descriptors Yes Light conditions Yes 7 descriptors Yes 7 descriptors Yes Yes Reported crashes Not specified All severities All injury severities All severities All severities All severities Definition of non-fatal Severe and non- A: Suspected serious injury Injured, admitted to Serious: Requiring Hospitalized, injured Severe severe injuries B: Suspected minor injury hospital Injured, required medical treatment Non-hospitalized, Slight C: Possible injury medical treatment Minor: Other injuries injured Damage Fatalities Within 30 days Within 30 days Within 30 days Within 30 days Within 30 days Within 30 days Link with hospital data No No In Western Australia No Yes No Speed limit Yes Yes Yes Yes Yes Yes Surface conditions Yes 10 descriptors Yes 3 descriptors 9 descriptors 3 descriptors Road curve No Yes Yes 4 descriptors 5 descriptors Yes Road segment gradient No Yes No No No No Age Yes Date of birth Yes Yes Yes Yes Gender Yes Yes Yes Yes Yes Yes Nationality Yes No Foreign drivers identified Foreign drivers identified Yes Yes Injury status No 5 descriptors 4 descriptors Yes 5 descriptors Yes Driver action No 19 descriptors In crash narrative In crash narrative 23 descriptors Yes Pedestrian action No 11 descriptors In crash narrative In crash narrative 11 descriptors Yes Violation codes No Yes Yes Yes No No Safety equipment Yes Yes Yes Yes Yes Yes Seating position No Yes Yes Yes Yes Yes ADT No Yes No Yes No No Curve radius No Yes No Yes No No Length No Yes No Yes No No Source. Table adapted from Montella et al. (2013) and Casado-Sanz et al. (2019). a b Directive 2008/96/EC of the European Parliament and of the Council of 19 November 2008 on road infrastructure safety management, Minimum Uniform Crash Criteria Model (MMUCC). Guideline c d Model minimum uniform crash criteria. National Highway Traffic Safety Administration (NHTSA), only pedestrian and cyclist ages in coded crash listing. Other ages in police crash reports, speed limit recorded since 2015, average daily traffic. Darus et al. 5 ββiXn independent variables as all the VIF values were less than 2. P = ββiXn ni (3) Second, the Small-Hsiao test of the IIA assumption was ∑∀II ′∈ examined (Table 5). The results reject the null hypothesis. It where β is a vector of coefficient that can be estimated for can be concluded that the IIA assumption has not been vio- severity level i using standard maximum likelihood methods. lated. In the next step, multinomial logistic regression analy- As the explanatory variables do not vary across injuries, the sis was performed using a stepwise backward elimination I−1 log-odd ratios of the model become: method to estimate the model. Table 6 adopted a design by Tay et al. (2011) and slight injury collision was selected as ni the reference category based on previous studies (Tay et al., ln == ββ xx −= ββ − xi ,, =… 11 …= I () in In iI n (4) ni 2011). As demonstrated in Table 6, the likelihood ratio chi- square of 302.573 with a small p-value of <.0001 demon- Only the difference in coefficient is identifiable, therefore, strates that the model fits the data significantly. The results the coefficient is identifiable up to an additive constant. The demonstrate that 11 explanatory variables had significant coefficients of one outcome (the base case) are set to zero to relationships between specific road and traffic environment resolve inter determinacy (Carson & Mannering, 2001). The characteristics and the injury severity level of child-pedes- slight injury collisions are used as a base case following trian collisions. The variables with a significant level at (Amoh-Gyimah et al., 2017; Tay et al., 2011). α = .1, α = .05, and α = .01 were included (Pour-Rouholamin As the MNL is a nonlinear model, the estimated coeffi- & Zhou, 2016; Rifaat et al., 2011; Tay et al., 2008). It should cients of the explanatory variables (independent variables) be noted that some variables were retained, although only do not represent their effect on the dependent variables. So, one category in the same factor demonstrated a significant the relative risk ratio (RRR) was computed relative to the value (Kockelman & Kweon, 2002; Tay et al., 2011). The base category to show the effect of significant risk factors. effects of each significant variable were interpreted using the From equation (4), the relative probability of an injury acci- estimated coefficients and relative risk ratios (RRR). dent (i = 2) to the base case (i = 1) is: With regard to geometry segment characteristics, cross Pr i = 2 2 () intersections and T/Y intersections lead to severe and fatal xβ() (5) injury collisions. Serious injury collisions are 1.8 times Pr i = 1 () more likely to occur at the cross intersection (p = .084, RRR = 1.819) than slight injury collisions. On the other The relative risk ratio (RRR) for binary variables is written hand, child-pedestrians who were hit at the T/Y intersection as follows: are 1.7 times more likely to result in fatal injury collisions (6) RRR = e (p = .042, RRR = 1.749). Generally, the RRR refers to an increase if the RRR is more The estimate findings showed that accidents involving seri- than 1 (RRR > 1) or a decrease if the RRR is less than 1 ous injuries occurred at the cross intersection (four-legged (RRR < 1) in the risk of a specific severity level of collision intersection) and the T/Y intersection (three-legged intersec- compared to the base case (Çelik & Oktay, 2014; Rifaat tion). In a study conducted in New York by Ukkusuri et al. et al., 2011). SPSS version 23 was used to estimate the MNL (2012), it was discovered that four-legged and five-legged and the RRR. intersections were linked with an increase in collision fre- quency. This is consistent with the fact that a larger number of legged intersections results in a greater number of conflict Results and Discussion sites, implying a greater probability of collisions at intersec- From the total of 2,518 vehicle collisions involving child- tions (Lee et al., 2016). Ewing and Dumbaugh (2009) also pedestrians, 14.5% were classified as fatal injury collisions, noted that vehicle-vehicle and vehicle-pedestrian accidents 30.1% were classified as serious injury collisions, and 55.4% occur mostly near intersections. Additionally, several studies were classified as slight/minor injury collisions. Table 3 have demonstrated that the density of road intersections may presents the distribution of child-pedestrian collisions based also have a traffic impact on child-pedestrians. For example, on 16 risk factors and 3 injury severity levels. previous scholars (Dissanayake et al., 2009; Steinbach et al., 2010) revealed that the higher density of road intersections was associated with higher child-pedestrian casualties. In a Multinomial Logit Model recent study conducted in Ulsan, Korea, Lee et al. (2016) dis- The estimation of the multinomial logit model is displayed in covered that the population density, number of marked cross- Table 6. First, multicollinearity was tested during the vari- walks, main road width, and number of building entrances at able selection process. There was no multicollinearity issue the intersection were associated with the occurrence of child- in the current study (Table 4). Multicollinearity among the pedestrian collisions at intersections. independent variables was checked using the VIF test. It was The severity of injury collisions is significantly associ- found that there was no multicollinearity issue among the ated with the types of road surface materials on collision 6 SAGE Open Table 3. Distribution (%) of Collision Severity and Contributing Factors. Injury severity levels of collisions (%) Variables Slight/minor Serious Fatal Total Road geometry Bend 5.4 5.9 6.0 5.6 Roundabout 0.2 0.1 0.3 0.2 Cross junction 1.4 2.2 0.3 1.5 T/Y junction 4.2 3.4 6.0 4.2 Staggered junction 0.1 0.1 0.3 0.2 Interchanges 0 0 0.3 0 Straight 88.7 88.1 86.9 88.3 Road surface material Crusher run 2.9 1.3 3.6 2.5 Brick 6.1 5.4 6.3 5.9 Concrete 0.4 0.7 1.1 0.6 Earth 0.5 0.7 3.0 0.9 Bitumen 90.1 91.9 86.1 90.1 Road surface defect Surface defect 2.2 2.0 3.3 2.3 Good condition 97.8 98.0 96.7 97.7 Shoulder type Paved 41.9 38.2 41.3 40.7 Unpaved 58.1 61.8 58.7 59.3 Road surface condition Wet 2.7 4.8 4.9 3.7 Others (oily, sandy, flood, construction works) 1.1 0.9 1.4 1.1 Dry 96.2 94.3 93.7 95.3 Traffic system One-way traffic (single carriageway) 15.9 13.3 15.6 15.1 Two-way traffic (Three lanes divided with bidirectional flow) 1.1 1.8 3.0 1.6 Dual carriageway (Multilane divided with bidirectional flow) 1.3 2.6 5.7 2.3 Two-way traffic (undivided with bidirectional flow) 81.6 82.2 75.7 80.9 Road marking category Double lane line (overtaking not permissible) 9.7 9.6 10.1 9.7 One-way lane line marking 4.9 2.8 3.8 4.1 Divider line marking 5.2 5.0 8.7 5.7 U-turn 0.1 0 0.3 0.1 No marking 24.4 20.7 15.6 22.0 Single lane line (overtaking permissible) 55.8 61.8 61.5 58.4 Control type Police 0.8 0.3 2.2 0.8 Other agencies 0.9 1.1 1.4 1.0 Traffic light 1.8 0.8 0.5 1.3 Pedestrian crossing 1.6 1.6 1.4 1.5 Pedestrian crossing with traffic light 0.4 0.7 0.5 0.5 Level crossing 0.0 0.1 0 0 Yellow line 2.9 1.8 3.3 2.7 Yellow box 1.9 0.9 0.5 1.4 No control 89.6 92.7 90.2 90.6 Speed limit 50 km/h 29.7 24.3 17.5 26.3 70 km/h 17.2 16.8 21.9 17.8 80 km/h 3.8 6.5 9.6 5.4 (continued) Darus et al. 7 Table 3. (continued) Injury severity levels of collisions (%) Variables Slight/minor Serious Fatal Total 90 km/h 7.9 14.3 15.3 10.9 110 km/h 0.5 0.1 0 0.3 Others 40.9 38.0 35.8 39.3 Time collision 0700 to 0959 8.3 8.8 12.1 9.0 1600 to 1859 31.3 32.8 23.3 30.6 1900 to 2459 14.3 16.8 18.8 15.7 0100 to 0659 3.2 5.1 7.0 4.3 1000 to 1559 42.9 36.6 38.8 40.4 Day of week Weekend 25.7 24.0 27.0 25.4 Weekday 74.3 76.0 73.0 74.6 Weather condition Windy 0.4 0.1 0.3 0.3 Foggy 3.9 5.7 6.8 4.9 Rain 2.4 3.3 3.3 2.8 Clear 93.3 90.9 89.6 92.0 Lighting condition Dawn/Dusk 6.3 7.5 7.4 6.8 Dark with street light 8.2 8.3 14.2 9.1 Dark without street light 3.4 5.7 7.1 4.6 Day 82.1 78.5 71.3 79.4 Type of location City 6.2 4.1 3.0 5.1 Urban 12.0 9.4 15.6 11.8 Built-up area 17.6 15.7 12.8 16.4 Rural 64.2 70.8 68.6 66.8 Land use Office 0.019 0.949 1.019 Shopping −0.148 0.525 0.863 Industrial/construction −0.464 0.303 0.629 School 16.1 14.1 6.8 14.2 Others 47.5 56.3 58.2 51.7 Residential 23.7 20.2 25.4 22.9 Hit and run Yes 9.5 8.2 9.0 9.1 No 90.5 91.8 91.0 90.9 Note. Location type: Metropolitan/city-Population more than 75,000 (Elsayed, 2012). Urban area-Gazetted areas with their adjoining built-up areas, which had a combined population of 10,000 or more at the time of the Census 2010 or the special development area that can be identified, which at least had a population of 10,000 with at least 60% of population (aged 15 years and above) were involved in non-agricultural activities (Department of Statistics Malaysia, 2020), Built-up areas were contiguous to a gazetted area and had at least 60% of their population (aged 15 years and above) engaged in non- agricultural activities (Department of Statistics Malaysia, 2020), Rural area is an area with a population less than 10,000 people, having agriculture and natural resources either clustered, linear, or scattered (Hamid & Toyong, 2014). severity and three types of road surface materials. The esti- earth surfaces (RRR = 9.697), respectively. The findings mation results indicated that serious injury collisions are less showed that the presence of concrete and the ground surface likely to occur on crusher run surfaces (p = .041, RRR = 0.471) enhanced the likelihood of sustaining a fatal injury. The in comparison to bitumen. Meanwhile, child-pedestrians crusher run surfaces may serve as a warning to drivers to who were hit by vehicles on concrete (p = .027) and earth slow down, thus reducing the effect of collisions on young- surfaces (p < .001) significantly resulted in fatal injury colli- sters on the road. While the flat surfaces of concrete and dirt sions. The fatal collisions are nearly 4.4 times (RRR = 4.435) result in greater impact collisions. These instances may be and 9.7 times more likely to occur on concrete surfaces and associated with the function of surface roughness. 8 SAGE Open Table 4. Results of the Multicollinearity Test. Standardized coefficients Collinearity statistics Beta T stat Tolerance VIF (Constant) 6.219 Road geometry –.007 –.336 .984 1.017 Road surface material .008 .402 .980 1.020 Road surface defect –.018 –.893 .970 1.031 Shoulder type .007 .312 .897 1.114 Road surface condition .057 2.686 .864 1.158 Traffic system .051 2.522 .968 1.033 Control type –.013 –.637 .961 1.041 Day of week .003 .164 .992 1.008 Weather condition –.048 –2.256 .870 1.149 Lighting condition .052 2.379 .818 1.222 Type of location –.062 –2.945 .892 1.121 Land use .015 .731 .931 1.074 Hit and run .011 .561 .990 1.010 Table 5. Small-Hsiao Test of IIA Assumption. 2 2 lnL (full) lnL(omit) Chi df p > chi Evidence Slight –448.2 –310.785 275.0 54 0.000 Against Ho Serious –1,056.7 –388.9 1,335.7 54 0.000 Against Ho Fatal –1,527.4 –667.4 1,720.2 54 0.000 Against Ho Note. A significant test is evidence against Ho. Additionally, prior research established that the roughness related to all age groups. This could be due to wet road sur- level was strongly associated with the frequency of vehicle faces affecting tire-pavement skid resistance, which is one of collisions and the degree of injuries (Anastasopoulos & the important factors in controlling vehicle direction, speed Mannering, 2009, 2011). Nevertheless, no literature was and short braking distances (Kokkalis & Panagouli, 1998), found on road surface conditions related to the child-pedes- and might have a higher collision impact on child-pedestri- trian severity collision. ans on roads. Other studies confirmed that friction between Overall, the association between severity levels of colli- tires and wet roads had a significant effect on vehicle colli- sions and road surface conditions was found to be statisti- sions (Najafi et al., 2017), as well as ineffective braking sys- cally significant. The results of the current study indicate that tems, causing vehicles to have less control (Graham & the risk of serious injury collisions generally increases on Glaister, 2003). wet road surfaces. The occurrence of serious injury colli- The impact of traffic system categories was also exam- sions is 1.6 times more likely to occur in wet surface condi- ined in this study. It should be noted that the classification of tions (p = .044, RRR = 1.645) compared with slight injury the traffic system category level is adopted from “Form collisions. Similarly, previous studies have reported the POL27” designed by the Royal Malaysian Police. Every cat- effects of wet road surfaces on child-pedestrian collisions. egory level varies with regard to the number of lanes, the For example, Koopmans et al. (2015) demonstrated that wet type of median (divided or undivided), or the type of flow surface conditions were significantly associated with child- (bidirectional or unidirectional). The findings indicated that pedestrian (aged ≤19 years old) fatalities. Meanwhile, a collisions occurring on three lanes (two-way divided) and study done in South Africa by Hobday and Knight (2010) dual carriageways (divided) are positively significant with showed that the highest percentage of fatal injury collisions both serious (three lanes, p = .07 and dual carriageways, involving child-pedestrians (<15 years old) occurred on p = .014) and fatal injury collisions (three lanes, p = .048 and freeways involving buses and trucks in wet surface condi- dual carriageways, p = .001) with reference to the two-way tions. In addition, the relative risk of far-side accidents had a (undivided) system. Serious injury collisions are two times significant effect when the road surface was wet, as reported more likely to occur on three lanes (RRR = 2.046) and 2.3 by Dunbar (2012). This exploratory variable was tested for a times (RRR = 2.327) on dual carriageways. Moreover, child- specific age group and it was found that it was significantly pedestrians are 2.5 times and 4.1 times more likely to sustain 9 Table 6. Multinomial Logit Model of Child-Pedestrian Collisions Severity. Number of observations: 2,518 Log-likelihood at zero: 3,234.843 Log-likelihood at convergence: 3,537.416 Chi-square: 302.573 p-value < .001 Base case: Slight crash injuries Serious crash injuries Fatal crash injuries Category Explanatory variable Reference B Sig Relative risk ratio B Sig Relative risk ratio Intercept –0.878 0.000 –1.647 0.001 Geometry Road geometry Straight Bend 0.022 0.911 1.023 0.018 0.947 1.018 Roundabout –0.543 0.643 0.581 0.475 0.696 1.609 Cross junction 0.598 0.084* 1.819 –1.404 0.176 0.246 T/Y junction –0.128 0.604 0.880 0.559 0.042** 1.749 Staggered junction 0.150 0.904 1.162 0.915 0.470 2.498 Interchanges 1.499 – 4.475 23.206 0.998 1.1E Road surface material Bitumen Crusher run –0.753 0.044** 0.476 0.258 0.472 1.294 Brick –0.104 0.609 0.902 0.030 0.906 1.031 Concrete 0.704 0.253 2.022 1.502 0.026** 4.489 Earth 0.382 0.527 1.466 2.214 0.000*** 9.155 Road surface condition Dry Wet 0.498 0.044** 1.645 0.415 0.183 1.514 Others –0.051 0.916 0.950 –0.160 0.787 0.852 Traffic Traffic system Two-way traffic (undivided with bidirectional flow) One-way traffic (single carriageway) 0.078 0.603 1.081 0.225 0.243 1.253 Two-way traffic (Three lanes divided 0.716 0.070* 2.046 0.900 0.048** 2.461 with bidirectional flow) Dual carriageway (Multilane divided 0.845 0.014** 2.327 1.414 0.001*** 4.112 with bidirectional flow) Road marking category Single lane line (overtaking permissible) Double lane line (overtaking not 0.023 0.891 1.024 0.261 0.239 1.299 permissible) One-way lane line marking –0.506 0.070* 0.603 –0.428 0.219 0.652 Divider line marking –0.226 0.304 0.798 0.084 0.742 1.087 –7 U-turn –14.911 0.995 3.3 × 10 1.080 0.512 2.945 (continued) 10 Table 6. (continued) Base case: Slight crash injuries Serious crash injuries Fatal crash injuries Category Explanatory variable Reference B Sig Relative risk ratio B Sig Relative risk ratio No marking –0.135 0.291 0.874 –0.559 0.004*** 0.572 Control type No control Police –1.148 0.140 0.317 0.858 0.096* 2.359 Other agencies 0.336 0.475 1.399 0.338 0.576 1.402 Traffic light –0.858 0.073* .424 –2.220 0.035** 0.109 Pedestrian crossing 0.131 0.731 1.140 0.099 0.860 1.104 Pedestrian crossing with traffic light 0.651 0.295 1.918 0.757 0.395 2.132 Level crossing –15.916 0.995 8.3 × 10 –0.441 1.000 0.644 Yellow line –0.420 0.198 0.657 0.470 0.198 1.600 Yellow box –0.663 0.130 0.515 –1.213 0.110 0.297 Speed limit 50 km/h 70 km/h 0.052 0.727 1.055 0.508 0.011** 1.662 80 km/h 0.571 0.012** 1.754 1.175 0.000*** 3.238 90 km/h 0.632 0.000*** 1.844 0.851 0.000*** 2.342 –7 110 km/h –1.222 0.261 0.292 –13.910 0.990 9.1x10 Others 0.063 0.599 1.060 0.366 0.039** 1.442 Temporal trend Time collision 1000 to 1559 0700 to 0959 0.190 0.284 1.209 0.360 0.097* 1.433 0000 to 0659 0.191 0.093* 1.210 0.639 0.021** 1.895 1600 to 1859 0.299 0.119 1.348 –0.310 0.058* 0.734 1900 to 2459 0.490 0.035** 1.632 –0.285 0.268 0.752 Environmental Lighting condition Day Dawn/Dusk 0.041 0.835 1.041 0.180 0.493 1.198 Dark with street light –0.077 0.721 0.926 0.771 0.003*** 2.162 Dark without street light 0.287 0.275 1.332 0.883 0.006*** 2.418 Land use Type of location Rural characteristic City –0.486 0.041** 0.615 –0.915 0.010*** 0.400 Urban –0.250 0.130 0.779 0.154 0.439 1.166 Built-up area –0.063 0.636 0.938 –0.227 0.231 0.797 Land use Residential Office 0.019 0.949 1.019 –0.418 0.298 0.658 Shopping –0.148 0.525 0.863 –0.871 0.014** 0.418 Industrial/construction –0.464 0.303 0.629 0.554 0.179 1.740 Bridge 0.212 0.668 1.236 –1.559 0.148 0.201 School 0.039 0.817 1.040 –1.135 0.000*** 0.321 Others 0.173 0.116 1.188 –0.102 0.521 0.903 *Significant at the 90% confidence interval, **significant at the 95% confidence interval, and ***significant at the 99% confidence interval. Slight injury crash used as a base case or reference. Darus et al. 11 fatal injuries if they are hit on three lanes (RRR = 2.461) and previous studies (Çelik & Oktay, 2014; Lee & Abdel-Aty, dual carriageways (RRR = 4.112). The results indicate that 2005; Rifaat et al., 2011). In the presence of traffic signals, child-pedestrians are at a greater risk of fatality and serious drivers might be more cautious and tend to slow down, injuries in traffic system categories with multilane roadways thereby resulting in a lower impact of injury collisions (Lee (more than two lanes), divided medians, and bidirectional & Abdel-Aty, 2005; Rifaat et al., 2011). Nevertheless, fatal traffic flow (two-way). This result is in agreement with a injury collisions (p = .086, RRR = 2.430) are 2.4 times more study conducted in Hong Kong (Sze & Wong, 2007) and likely to occur at a location controlled by the police. No ref- Riyadh (Al-Ghamdi, 2002). Besides, Wong et al. (2007) erence was found related to the location controlled by the found that two-way and multi/dual carriageways were sig- police. nificantly related to an increased probability of severe injury Additionally, the effect of various kinds of speed limit levels. While, Al-Ghamdi (2002), reported that two-way indications was investigated, with roads having a speed limit divided roadways (two-way with median) are the most fre- indication of 70 km/h (p = .007, RRR = 1.726) being substan- quent locations for pedestrian-vehicle collisions in Riyadh. tially linked with fatal injuries. Additionally, highways with Nonetheless, several studies examined each characteristic of speed limit signs of 80 and 90 km/h were strongly linked the traffic system associated with severity level separately. with severe and fatal injury accidents. Additionally, acci- For example, several studies (Agran et al., 1996; Mueller dents involving speeds of 80 km/h (p = .001, RRR = 3.417) et al., 1990) found that roadways with more than two travel and 90 km/h (p = .001, RRR = 2.347) are 3.4 times and 2.3 lanes and wider lanes are related to higher risks of child- times more likely to result in fatal injuries, respectively. This pedestrian collisions. The results indicate that wider lanes result indicates that highways having posted speed limits of may be correlated with higher traffic speed and lead to higher 70 km/h more are linked with a higher severity level. occurrence rates and risks of child-pedestrian collisions Similarly, prior research has shown that locations with higher (Abdel-Aty et al., 2007; Al-Ghamdi, 2002; Ewing & speed limit signs are associated with higher impact speeds Dumbaugh, 2009). Nevertheless, Abdel-Aty et al. (2007), and a larger probability of fatal collisions (Davis, 2001). revealed that the higher collision involvement among school- Additionally, Davis (2001) created a discrete outcome model aged children in Florida increases as the number of lanes that links the degree of an injured pedestrian’s injury to the increases on divided roads. This study indicated that a multi- hitting vehicle’s speed. Additionally, it was shown that child- lane roadway with a median presence influences the risk of pedestrians aged 0 to 14 years old are more likely to have collisions. With regard to the type of flow, Al-Ghamdi (2002) severe injuries when collision speeds exceed 40 km/h. found that high levels of injuries were associated with two- Additionally, if they are struck at a speed greater than way roadways with a median. Therefore, the risk of injury 75 km/h, fatal injuries are quite probable. collisions may be caused by any characteristics of the traffic With respect to the time of collisions, the findings revealed system, which are the number of lanes, the type of median that collisions occurring from 000 to 0659 hours (midnight to (divided or undivided), and the type of flow (bidirectional or early morning) were significantly related to serious (=0.093, unidirectional), or any combination of the aforementioned RRR = 1.210) and fatal injury collisions (p = .021, characteristics. RRR = 1.895). Also, collisions are likely to be fatal (p = .097, As for the road marking category, serious (p = .07, RRR = 1.433) from 0700 to 0959 hours (morning). While, RRR = 0.603) collision injuries were found to be less likely serious injury collisions are likely to occur from 1900 to to occur on roads with a one-way line. Also, fatal (p = .004, 2459 hours (night). Nevertheless, collisions are less likely to RRR = 0.572) collision injuries are less likely to occur on cause fatal injury collisions from 1600 to 1859 hours (late roads with no marking lines. As such, serious injury colli- afternoon). This finding implies that child-pedestrians in sions are 1.7 times (1/0.603) and fatal injury collisions are Malaysia experienced more severe injury collisions during almost 1.7 times (1/0.572) more likely to occur on roads with the early morning, morning, and night time. This may be a single line compared to roads with a one-way line and no related to peak traffic periods and school opening and clos- marking line, respectively. This result might be due to the ing times, as observed in previous studies (Pitt et al., 1990; fact that vehicles are allowed to overtake other vehicles on Yiannakoulias et al., 2002). Nevertheless, fatal collisions roads with a single line, which can lead to higher severity of during midnight might be due to negligence factors by other collisions. For example, Leden et al. (2006) found that child- road users. Moreover, fatigue, tiredness, stress, speeding, pedestrians in Finland were frequently involved in collisions and lack of visibility at night may be attributed to driver fac- involving overtaking vehicles at mid-block. With regard to tors (Amoh-Gyimah et al., 2017). traffic control devices, the results of the current study dem- The natural lighting conditions were found to be signifi- onstrate that collisions at traffic lights were found to be less cant with fatal injury. Dark conditions with street lights likely to result in serious (p = .073, RRR = 0.424) and fatal (p = .003, RRR = 2.162) are positively associated with fatal injury collisions (p = .035, RRR = 0.109). This finding indi- injury collisions. Additionally, fatal injury collisions are 2.4 cated that the presence of traffic lights reduced the probabil- times more likely to occur in the dark without street light ity of serious and fatal injuries, which was in agreement with (p = .006, RRR = 2.418) compared to slight injury collisions. 12 SAGE Open This result indicates that child-pedestrians are at a higher et al. (2007) found that the presence of a driveway and turn- injury risk in darkness. This result is consistent with previous ing bay decreases the incidence rates and severity of injuries findings (Amoh-Gyimah et al., 2017; Koopmans et al., 2015; in school areas. Nevertheless, this finding contradicts Abdel- Mohamed et al., 2013; Rifaat et al., 2011), which demon- Aty et al. (2007), whose study demonstrated that school chil- strated that darkness may affect the visibility of drivers and dren are at greater risk due to the fact that the middle and pedestrians. Rifaat et al. (2011) highlighted that drivers may high schools in Orange County, Florida, tend to be located struggle to identify hazards and other road users. Additionally, near multi-lane high-speed roads. drivers may be unable to perform appropriate evasive maneu- vers and slow down, resulting in more severe accidents. Conclusion In relation to location characteristics, the results reveal that city areas were less likely to have serious (p = .041, The current study identified several risk factors associated RRR = 0.615) or fatal injury collisions (p = .010, RRR = 0.400) with the increasing severity of vehicle collisions involving compared to rural areas. This result indicates that child- child-pedestrians. The explanatory variables associated with pedestrians in urban areas may be exposed to fewer severe increased probability of fatal injury collisions include: t/y injuries compared to those who live in rural areas. This find- intersection; concrete and earth-road surfaces; two-way traffic ing is similar to a few other studies (Afukaar et al., 2003; and dual carriageways; posted speed limits of 70 to 90 km/h; Doukas et al., 2010; Singh et al., 2016). Evidence indicated time of collision: 0 to 0659 hours (early morning) and 0700 to that lack of road maintenance, poor quality of public trans- 0959 hours (morning); lighting conditions, including dark port, low awareness of traffic rules, higher vehicle speeds, without street lights and dark with street lights; and control fewer separated pedestrian facilities, and limited access to type involving police. Also, cross intersections, posted speed medical emergency services might be contributing factors to limits of 80 to 90 km/h, and time of collision, which are from severe injuries in rural areas (Afukaar et al., 2003; Singh 0 to 0659 hours (early morning) and from 1900 to 2459 hours et al., 2016). This finding shows that the specific carriage- (night), significantly increased serious injury collisions. way features in different locations will impact the road From a traffic engineering perspective, road characteris- accident severity levels. Alternatively, past researchers tics, road designs, and traffic operations play an important (Yiannakoulias et al., 2002) also suggested that the associa- role in providing safe walking conditions for vulnerable road tion between physical environmental characteristics in com- users (Oxley et al., 2018). Fundamentally, child-pedestrians bination with spatial analysis with different analytical should be provided with good and protective facilities based approaches could be adopted for a comprehensive overview on their abilities (Assailly, 1997) and skills. For example, of the influence of accident location on severity levels. there should be special attention given to locations with com- Besides that, the effect of different types of land use char- plex road geometrics and traffic systems, such as t/y intersec- acteristics was examined. From the model estimation, it was tions, cross intersections, and multilane roads with found that slight crash injuries (p = .014, RRR = 0.418) are bidirectional flows, as all these locations tend to be more nearly 2.4 times (1/0.418) more likely to occur in shopping severe. The reduction in child-pedestrian collisions is associ- areas compared to fatal crash injuries. Similar to this finding, ated with lighting conditions and road surface conditions. It Clifton and Kreamer-Fults (2007) found that commercial should be noted that child-pedestrian collisions may be land uses near public schools were associated with higher attributed to drivers’ negligence and irresponsible behavior. pedestrian-vehicular collision severity. Meanwhile Elias and Importantly, dark areas should be installed with adequate Shiftan (2014) and Mohamed et al. (2013), reported that chil- street lighting to increase visibility. Furthermore, road safety dren who live in areas of mixed land use (including commer- educational programs and campaigns should also be targeted cial land use) are significantly exposed to road collisions. at drivers to increase their awareness of pedestrians’ and Clifton and Kreamer-Fults (2007) revealed that the location cyclists’ activities (Desapriya et al., 2011; Tay et al., 2011). characteristics were likely associated with higher levels of Moreover, modifications to the physical environment should pedestrian demand and thus high absolute numbers of be combined with effective training techniques involving crashes. Elias and Shiftan (2014) demonstrated that metered children, especially in school zones, to reduce the selection parking facilities, which are located in commercial areas of poor routes (Schwebel et al., 2012). Notably, future where speeds tend to be lower, have a significant effect on research should examine the human factors concerning chil- reducing fatality risks. Also, it was found that slight crash dren’s behavior and development in order to better under- injuries (p = .000, RRR = 0.321) are nearly 3.1 (1/0.321) stand child-pedestrians’ abilities and limitations in designing times more likely to occur in school areas. This finding is pedestrian facilities. similar to other studies (Clifton & Kreamer-Fults, 2007; Pitt et al., 1990). Pitt et al. (1990) found that the lower incidence Limitation rates and lower severity of injuries in the vicinity of schools might be correlated to the success of road safety education There are a few limitations to this research. To begin, data on programs and other traffic safety measures, while Clifton physical environmental features was recorded separately Darus et al. 13 from data on human aspects (age, gender, behavior, etc.). As Amoh-Gyimah, R., Aidoo, E. N., Akaateba, M. A., & Appiah, S. K. (2017). The effect of natural and built environmen- a result, association between human factors variables and tal characteristics on pedestrian-vehicle crash severity in physical environmental features has not been examined. Ghana. International Journal of Injury Control and Safety Second, the database does not include spatial features such Promotion, 24(4), 459–468. https://doi.org/10.1080/174573 as the dimensions of pedestrian pathways, the layout and 00.2016.1232274 usage of the area between the road and the sidewalk. It is Anastasopoulos, P. C., & Mannering, F. L. (2009). A note on mod- recommended that the analysis of physical environment eling vehicle accident frequencies with random-parameters characteristics in combination with spatial analysis with dif- count models. Accident Analysis and Prevention, 41(1), ferent analytical approach should be further investigated. 153–159. https://doi.org/10.1016/j.aap.2008.10.005 Another drawback of the research was that the database was Anastasopoulos, P. C., & Mannering, F. L. (2011). An empirical only available until 2014. In terms of future work, it may be assessment of fixed and random parameter logit models using prudent to do further analysis of current data in order to crash- and non-crash-specific injury data. Accident Analysis and Prevention, 43(3), 1140–1147. https://doi.org/10.1016/j. enhance the predictability of the MNL models. Despite data aap.2010.12.024 limitations, these statistics offer sufficient information on Assailly, J. P. (1997). Characterization and prevention of child road and environmental factors for preventive measures. pedestrian accidents: An overview. Journal of Applied Developmental Psychology, 18(2), 257–262. https://doi.org Acknowledgments /10.1016/S0193-3973(97)90039-3 The authors wish to thank Malaysian Royal Police (RMP) and Barton, B. K., Ulrich, T., & Lyday, B. (2012). The roles of gender, Malaysian Institute of Road Safety Research (MIROS) for provid- age and cognitive development in children’s pedestrian route ing road accident data for this project and financial support from selection. Child Care Health and Development, 38(2), Ministry of Education, Malaysia. 280–286. https://doi.org/10.1111/j.1365-2214.2010.01202.x Carson, J., & Mannering, F. (2001). The effect of ice warning signs Declaration of Conflicting Interests on ice-accident frequencies and severities. Accident Analysis and Prevention, 33(1), 99–109. https://doi.org/10.1016/S0001- The author(s) declared no potential conflicts of interest with respect 4575(00)00020-8 to the research, authorship, and/or publication of this article Casado-Sanz, N., Guirao, B., Galera, A. L., & Attard, M. (2019). Investigating the risk factors associated with the severity of the Funding pedestrians injured on Spanish crosstown roads. Sustainability, The author(s) disclosed receipt of the following financial support 11(19), 1–18. https://doi.org/10.3390/su11195194 for the research, authorship, and/or publication of this article: This Çelik, A. K., & Oktay, E. (2014). A multinomial logit analy- research was funded by Ministry of Education, Malaysia research sis of risk factors influencing road traffic injury severi- Grant no. FRGS/1/2021/TK02/UKM/02/1. ties in the Erzurum and Kars provinces of Turkey. Accident Analysis and Prevention, 72, 66–77. https://doi.org/10.1016/j. aap.2014.06.010 ORCID iD Clifton, K. J., & Kreamer-Fults, K. (2007). An examination of Muhamad Nazri Borhan https://orcid.org/0000-0003-4517-5659 the environmental attributes associated with pedestrian- vehicular crashes near public schools. Accident Analysis References and Prevention, 39, 708–715. https://doi.org/10.1016/j. Abdel-Aty, M., Chundi, S. S., & Lee, C. (2007). Geo-spatial and aap.2006.11.003 log-linear analysis of pedestrian and bicyclist crashes involv- Clifton, K. J., Livi Smith, A. D., & Rodriguez, D. (2007). The ing school-aged children. Journal of Safety Research, 38(5), development and testing of an audit for the pedestrian envi- 571–579. https://doi.org/10.1016/j.jsr.2007.04.006 ronment. Landscape and Urban Planning, 80(1–2), 95–110. Afukaar, F. K., Antwi, P., & Ofosu-Amaah, S. (2003). Pattern of https://doi.org/10.1016/j.landurbplan.2006.06.008 road traffic injuries in Ghana: Implications for control. Injury Darus, N. S., Borhan, M. N., Ishak, S. Z., Ismail, R., & Mohd. Control and Safety Promotion, 10(1–2), 69–76. https://doi. Razali, S. F. (2018). Pattern of child pedestrian collisions and org/10.1076/icsp.10.1.69.14107 injuries in Malaysia. Jurnal Teknologi, 80(4), 1–8. https://doi. Agran, P. F., Winn, D. G., Anderson, C. L., Tran, C., & Del Valle, org/10.11113/jt.v80.11177 C. P. (1996). The role of the physical and traffic environment Davis, G. A. (2001). Relating severity of pedestrian injury to in child pedestrian injuries. Pediatrics, 98(6 Pt 1), 1096–1103. impact speed in vehicle-pedestrian crashes: Simple thresh- Ahmed, A., Farhan, A., Sadullah, M., & Yahya, A. S. (2020). old model. Transportation Research Record: Journal of the The effect of type of attributes on the fill-ability of acci- Transportation Research Board, 1773(1), 108–113. dent reporting forms (ARF). International Journal of Traffic Davis, N. K., & Rice, R. G. (1994). The role of the physical envi- and Transportation Management, 1(2), 11–18. https://doi. ronment in child pedestrian accidents. Journal of Advanced org/10.5383/jttm.01.02.002 Transportation, 28(2), 171–187. Al-Ghamdi, A. S. (2002). Pedestrian-vehicle crashes and analytical Department of Statistics Malaysia. (2020). Definition of urban area techniques for stratified contingency tables. Accident Analysis by Department of Statistic. https://www.dosm.gov.my/v1/ and Prevention, 34(2), 205–214. https://doi.org/10.1016/ index.php?r=column/cone&menu_id=bDA2VkxRSU40STcxd S0001-4575(01)00015-X kZ4OGJ0c1ZVdz09 14 SAGE Open Desapriya, E., Sones, M., Ramanzin, T., Weinstein, S., Scime, G., Kokkalis, A. G., & Panagouli, O. K. (1998). Fractal evaluation & Pike, I. (2011). Injury prevention in child death review: of pavement skid resistance variations. II: Surface wear. Child pedestrian fatalities. Injury Prevention, 17, 4–9. https:// Chaos, Solitons, and Fractals, 9(11), 1891–1899. https://doi. doi.org/10.1136/ip.2010.026914 org/10.1016/S0960-0779(98)00185-4 Dissanayake, D., Aryaija, J., & Wedagama, D. M. (2009). Koopmans, J. M., Friedman, L., Kwon, S., & Sheehan, K. (2015). Modelling the effects of land use and temporal factors on child Urban crash-related child pedestrian injury incidence and char- pedestrian casualties. Accident Analysis and Prevention, 41(5), acteristics associated with injury severity. Accident Analysis 1016–1024. https://doi.org/10.1016/j.aap.2009.06.015 and Prevention, 77, 127–136. https://doi.org/10.1016/j. Doukas, G., Olivier, J., Poulos, R., & Grzebieta, R. (2010). Exploring aap.2015.02.005 differential trends in severe and fatal child pedestrian injury in Leden, L., Gårder, P., & Johansson, C. (2006). Safe pedes- New South Wales, Australia (1997–2006). Accident Analysis trian crossings for children and elderly. Accident Analysis and Prevention, 42(6), 1705–1711. https://doi.org/10.1016/j. and Prevention, 38(2), 289–294. https://doi.org/10.1016/j. aap.2010.04.010 aap.2005.09.012 Dunbar, G. (2012). The relative risk of nearside accidents is high Lee, C., & Abdel-Aty, M. (2005). Comprehensive analysis for the youngest and oldest pedestrians. Accident Analysis of vehicle-pedestrian crashes at intersections in Florida. and Prevention, 45, 517–521. https://doi.org/10.1016/j. Accident Analysis and Prevention, 37, 775–786. https://doi. aap.2011.09.001 org/10.1016/j.aap.2005.03.019 Elias, W., & Shiftan, Y. (2014). Analyzing and modeling risk Lee, G., Park, Y., Kim, J., & Cho, G.-H. (2016). Association between exposure of pedestrian children to involvement in car crashes. intersection characteristics and perceived crash risk among Accident Analysis and Prevention, 62, 397–405. https://doi. school-aged children. Accident Analysis and Prevention, 97, org/10.1016/j.aap.2013.06.035 111–121. https://doi.org/10.1016/j.aap.2016.09.001 Elsayed, I. (2012). Effects of population density and land man- Mohamed, M. G., Saunier, N., Miranda-Moreno, L. F., & Ukkusuri, agement on the intensity of urban heat islands: A case study S. V. (2013). A clustering regression approach: A comprehen- on the city of Kuala Lumpur, Malaysia. In B. M. Alam (Ed.), sive injury severity analysis of pedestrian–vehicle crashes in Application of geographic information systems (pp. 267–282). New York, US and Montreal, Canada. Safety Science, 54, IntechOpen. 27–37. https://doi.org/10.1016/j.ssci.2012.11.001 Ewing, R., & Dumbaugh, E. (2009). The built environ- Mohamed, N., Wong, S. V., Hashim, H. H., & Othman, I. (2011). ment and traffic safety: A review of empirical evidence. An overview of road traffic injuries among children in Malaysia Journal of Planning Literature, 23, 347–367. https://doi. and its implication on road traffic injury prevention strategy org/10.1177/0885412209335553 (MRR 03/2011 Research report). Malaysian Institute of Road Graham, D. J., & Glaister, S. (2003). Spatial variation in road Safety Research. pedestrian casualties: The role of urban scale, density and Montella, A., Andreassen, D., Tarko A. P., & Turner, S. (2013). land-use mix. Urban Studies, 40(8), 1591–1607. https://doi. Crash databases in Australasia, the European Union, and the org/10.1080/0042098032000094441 United States. Transportation Research Record, 2386, 128–136. Grechkin, T. Y., Chihak, B. J., Cremer, J. F., Kearney, J. K., & https://doi.org/10.3141/2386-15 Plumert, J. M. (2013). Perceiving and acting on complex Mueller, B. A., Rivara, F. P., Lii, S.-M., & Weiss, N. S. (1990). affordances: How children and adults bicycle across two Environmental factors and the risk for childhood pedestrian- lanes of opposing traffic. Journal of Experimental Psychology motor vehicle collision occurrence. American Journal of Human Perception & Performance, 39(1), 23–36. https://doi. Epidemiology, 132(3), 550–560. org/10.1037/a0029716 Najafi, S., Flintsch, G. W., & Medina, A. (2017). Linking roadway Haddon, W. (1980). Advances in the epidemiology of injuries crashes and tire–pavement friction: A case study. International as a basis for public policy. Public Health Reports, 95(5), Journal of Pavement Engineering, 18(2), 119–127. https://doi. 411–421. org/10.1080/10298436.2015.1039005 Hamid, L. A., & Toyong, N. M. P. (2014). Rural area, elderly OECD. (2010). Reporting on serious road traffic casualties. http:// people and the house breaking crime. Procedia - Social and www.internationaltransportforum.org/irtadpublic/pdf/Road- Behavioral Sciences, 153, 443–451. https://doi.org/10.1016/j. Casualties-Web.pdf sbspro.2014.10.078 Oxley, J., Fildes, B., Ihsen, E., Charlton, J., & Day, R. (1997). Hashim, H. H., & Rahim, S. A. S. M. (2009, September 16–17). The Differences in traffic judgements between young and old construction of road accident analysis and database system in adult pedestrians. Accident Analysis and Prevention, 29(6), th Malaysia [Conference session]. The 4 IRTAD Conference, 839–847. https://doi.org/10.1016/S0001-4575(97)00053-5 Seoul, Korea. http://www.internationaltransportforum.org/irtad Oxley, J., Jamaludin, A., & Johnson, M. (2012). Analysis of child /Seoul2009.html pedestrian deaths and serious injuries in Malaysia. Journal of Hobday, M., & Knight, S. (2010). Motor vehicle colli- the Australasian College of Road Safety, 23(2), 30–37. sions involving child pedestrians in ethekwini in 2007. Oxley, J., O’Hern, S., Burtt, D., & Rossiter, B. (2018). Falling while Journal of Child Health Care, 14(1), 67–81. https://doi. walking: A hidden contributor to pedestrian injury. Accident org/10.1177/1367493509347059 Analysis and Prevention, 114, 77–82. https://doi.org/10.1016/j. Kockelman, K. M., & Kweon, Y. J. (2002). Driver injury sever- aap.2017.01.010 ity: An application of ordered probit models. Accident Analysis Peden, M. (2004). World report on road traffic injury prevention. and Prevention, 34(3), 313–321. https://doi.org/10.1016/ Public Health, 120(3), 280–280. https://doi.org/10.1016/j. S0001-4575(01)00028-8 puhe.2005.09.003 Darus et al. 15 Peek-Asa, C., & Zwerling, C. (2003). Role of environmental injury rates in London. Health & Place, 16(1), 34–42. https:// interventions in injury control and prevention. Epidemiologic doi.org/10.1016/j.healthplace.2009.08.002 Reviews, 25, 77–89. https://doi.org/10.1093/epirev/mxg006 Stevenson, M. R., Jamrozik, K. D., & Spittle, J. (1995). A case- Pitcairn, T. K., & Edlmann, T. (2000). Individual differences in control study of traffic risk factors and child pedestrian injury. road crossing ability in young children and adults. British International Journal of Epidemiology, 24(5), 957–964. https:// Journal of Psychology, 91(3), 391–410. doi.org/10.1093/ije/24.5.957 Pitt, R., Guyer, B., Hsieh, C. C., & Malek, M. (1990). The sever- Sze, N. N., & Wong, S. C. (2007). Diagnostic analysis of the ity of pedestrian injuries in children: An analysis of the pedes- logistic model for pedestrian injury severity in traffic crashes. trian injury causation study. Accident Analysis and Prevention, Accident Analysis and Prevention, 39(6), 1267–1278. https:// 22(6), 549–559. https://doi.org/10.1016/0001-4575(90)90027-I doi.org/10.1016/j.aap.2007.03.017 Pour-Rouholamin, M., & Zhou, H. (2016). Investigating the risk Tay, R., Choi, J., Kattan, L., & Khan, A. (2011). A multinomial factors associated with pedestrian injury severity in Illinois. logit model of pedestrian–vehicle crash severity. International Journal of Safety Research, 57, 9–17. https://doi.org/10.1016/j. Journal of Sustainable Transportation, 5, 233–249. https://doi. jsr.2016.03.004 org/10.1080/15568318.2010.497547 Rifaat, S. M., Tay, R., & de Barros, A. (2011). Effect of street pat- Tay, R., Rifaat, S. M., & Chin, H. C. (2008). A logistic model tern on the severity of crashes involving vulnerable road users. of the effects of roadway, environmental, vehicle, crash and Accident Analysis and Prevention, 43(1), 276–283. https://doi. driver characteristics on hit-and-run crashes. Accident Analysis org/10.1016/j.aap.2010.08.024 and Prevention, 40(4), 1330–1336. https://doi.org/10.1016/j. Rothman, L., Slater, M., Meaney, C., & Howard, A. (2010). Motor aap.2008.02.003 vehicle and pedestrian collisions: Burden of severe injury on Tiwari, G. (2020). Progress in pedestrian safety research. major versus neighborhood roads. Traffic Injury Prevention, International Journal of Injury Control and Safety 11(1), 43–47. https://doi.org/10.1080/15389580903452340 Promotion, 27(1), 35–43. https://doi.org/10.1080/1745730 Royal Malaysia Police. (2014). Laporan Perangkaan Kemalangan 0.2020.1720255 Jalan Raya Malaysia 2014. Polis DiRaja Malaysia. Toroyan, T. (2015). Global status report on road safety. World Runyan, C. W. (2003). Introduction: Back to the future – revisit- Health Orgainisation, p. 318. http://www.who.int/violence_ ing Haddon’s conceptualization of injury epidemiology and injury_prevention/road_safety_status/2015/en/ prevention. Epidemiologic Reviews, 25, 60–64. https://doi. Ukkusuri, S., Miranda-Moreno, L. F., Ramadurai, G., & Isa- org/10.1093/epirev/mxg005 Tavarez, J. (2012). The role of built environment on pedestrian Saidon, N., & Baguley, C. (1994, March). Road accidents in crash frequency. Safety Science, 50(4), 1141–1151. https://doi. Malaysia [Conference session]. Proceeding seminar on crash org/10.1016/j.ssci.2011.09.012 and injury control, Universiti Malaya, Kuala Lumpur. Ulfarsson, G. F., & Mannering, F. L. (2004). Differences in male Sandels, S. (1970). Young children in traffic. British Journal and female injury severities in sport-utility vehicle, minivan, of Educational Psychology, 40(2), 111–116. https://doi. pickup and passenger car accidents. Accident Analysis and org/10.1111/j.2044-8279.1970.tb02111.x Prevention, 36(2), 135–147. https://doi.org/10.1016/S0001- Schieber, R. A., & Thompson, N. J. (1996). Developmental risk 4575(02)00135-5 factors for childhood pedestrian injuries. Injury Prevention, Vinje, M. P. (1981). Children as pedestrians: Abilities and limi- 2(3), 228–236. https://doi.org/10.1136/ip.2.3.228 tations. Accident Analysis and Prevention, 13(3), 225–240. Schwebel, D. C., Davis, A. L., & O’Neal, E. E. (2012). Child pedes- https://doi.org/10.1016/0001-4575(81)90006-3 trian injury: A review of behavioral risks and preventive strat- Wazana, A., Krueger, P., Raina, P., & Chambers, L. (1997). A egies. American Journal of Lifestyle Medicine, 6, 292–302. review of risk factors for child pedestrian injuries: Are they https://doi.org/10.1177/0885066611404876 modifiable? Injury Prevention, 3(4), 295–304. https://doi. Shankar, V., Mannering, F., & Barfield, W. (1996). Statistical anal- org/10.1136/ip.4.3.243-b ysis of accident severity on rural freeways. Accident Analysis Whitebread, D., & Neilson, K. (2000). The contribution of visual and Prevention, 28(3), 391–401. https://doi.org/10.1016/0001- search strategies to the development of pedestrian skills by 4-11 4575(96)00009-7 year-old children. British Journal of Educational Psychology, Singh, D., Singh, S. P., Kumaran, M., & Goel, S. (2016). 70(4), 539–557. https://doi.org/10.1348/000709900158290 Epidemiology of road traffic accident deaths in children in Wong, S. C., Sze, N. N., & Li, Y. C. (2007). Contributory factors Chandigarh zone of North West India. Egyptian Journal of to traffic crashes at signalized intersections in Hong Kong. Forensic Sciences, 6, 255–260. https://doi.org/10.1016/j. Accident Analysis and Prevention, 39, 1107–1113. https://doi. ejfs.2015.01.008 org/10.1016/j.aap.2007.02.009 Sminkey, L. (2008). World report on child injury prevention. Injury Yao, J. F., Yang, J. K., & Fredriksson, R. (2006). Reconstruction Prevention, 14(1), 69. https://doi.org/10.1136/ip.2007.018143 of head-to-hood impact in an automobile-to-child-pedestrian Solagberu, B. A., Osuoji, R. I., Ibrahim, N. A., Oludara, M. A., collision. International Journal of Crashworthiness, 11(4), Balogun, R. A., Ajani, A. O., Idowu, O. E., Mustafa, I. A., 387–395. https://doi.org/10.1533/ijcr.2005.0118 & Sanni, F. O. (2014). Child pedestrian injury and fatality in Yiannakoulias, N., Smoyer-Tomic, K. E., Hodgson, J., Spady, D. a developing country. Pediatric Surgery International, 30(6), W., Rowe, B. H., & Voaklander, D. C. (2002). The spatial and 625–632. https://doi.org/10.1007/s00383-014-3516-8 temporal dimensions of child pedestrian injury in Edmonton. Steinbach, R., Green, J., Edwards, P., & Grundy, C. (2010). Race Canadian Journal of Public Health, 93(6), 447–451. http:// or place? Explaining ethnic variations in childhood pedestrian www.jstor.org/stable/41993563 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png SAGE Open SAGE

The Effect of Physical Environment Risk Factors on Vehicle Collisions Severity Involving Child-Pedestrians in Malaysia:

Loading next page...
 
/lp/sage/the-effect-of-physical-environment-risk-factors-on-vehicle-collisions-UuC4H0la88

References (92)

Publisher
SAGE
Copyright
Copyright © 2022 by SAGE Publications Inc, unless otherwise noted. Manuscript content on this site is licensed under Creative Commons Licenses.
ISSN
2158-2440
eISSN
2158-2440
DOI
10.1177/21582440211068494
Publisher site
See Article on Publisher Site

Abstract

This study is aimed at investigating the association between child-pedestrian severity levels of collisions and physical environmental variables. The outcome of this study could be applied to road safety intervention for improving engineering modifications related to children pedestrians. The retrospective analysis was carried out using 6-year data from Royal Malaysia Police records from the years 2009 to 2014. Multinomial logit modeling (MNL) was applied. The results demonstrated that the injury severity of the collisions is related to road geometry, road surface material, road surface condition, traffic system, road marking, traffic control type, lighting condition, speed limit, time of collision, type of location, and land use characteristics. Specifically, fatal injury collisions are significantly increased by t/y intersection; concrete and earth-road surfaces; two-way traffic and dual carriageways; posted speed limits of 70 to 90 km/h; time of collision: 0 to 0659 hours (early morning) and 0700 to 0959 hours (morning); lighting conditions, including dark without street light, and dark with street light; and control type involving police. Meanwhile, cross intersections’ posted speed limits of 80 km/h to 90 km/h and the time of collision from 0 to 0659 hours (early morning) and 1900 to 2459 hours (night), significantly increased serious injury collisions. Notably, the findings revealed the importance of more in-depth studies on physical environmental features that relate to child-pedestrians’ severity level of collisions. This is essential for improvements to physical environmental designs by policymakers. Thus, policymakers and stakeholders can utilize the findings to further improve the physical environment through structure and design. Keywords child-pedestrian, road injuries, collisions, severity children under the age of 10 are unable to engage with the Introduction traffic environment owing to their weak sensory and cogni- Child-pedestrians are considered to be one of the most vul- tive skills. Tiwari (2020) also highlighted that conventional nerable road users (VRUs) in non-motorized traffic groups pedestrian instruction is ineffective for young children, par- (Dissanayake et al., 2009). Child-pedestrians are highly ticularly those under the age of 10, whereas earlier research vulnerable to traffic injury due to their underdeveloped has shown a link between cognitive abilities and pedestrian physical, cognitive, developmental, behavioral, and sensory age (Barton et al., 2012; Pitcairn & Edlmann, 2000; Sandels, functions (Barton et al., 2012; Oxley et al., 1997; Schieber & 1970; Whitebread & Neilson, 2000). For example, Barton Thompson, 1996; Sminkey, 2008; Vinje, 1981). In general, child-pedestrians are still developing, resulting in a high Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia incidence of accidents and severe injuries. For instance, their Universiti Teknologi Mara, Shah Alam, Selangor, Malaysia diminutive size impairs their ability to see and be noticed by Malaysian Institute of Road Safety Research (MIROS), Kajang, Selangor, other road users (Oxley et al., 2012; Schieber & Thompson, Malaysia 1996; Solagberu et al., 2014). Consequently, this increases Corresponding Author: the risks of severe and fatal injury (Rothman et al., 2010; Yao Muhamad Nazri Borhan, Department of Civil Engineering/Sustainable et al., 2006). Urban Transport Research Centre (SUTRA), Faculty of Engineering & Growing research indicates that children’s cognitive Built Environment, Universiti Kebangsaan Malaysia, development has an effect on child-pedestrians’ safety. In a Bangi, Selangor 43600, Malaysia. Email: mnazri_borhan@ukm.edu.my study done by Sandels (1970), the author concluded that Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open et al. (2012) investigated the relationship between children’s road conditions, the number of traffic lanes, the volume of age, gender, the two elements of cognitive functioning (visual traffic, play areas, and buildings, are examples of a man- search and efficiency), and child-pedestrians’ route choices. made environment (Peek-Asa & Zwerling, 2003; Wazana The study discovered that males, particularly younger chil- et al., 1997). Meanwhile, weather, extreme temperatures, and dren, choose riskier pedestrian routes. Nonetheless, many lighting are part of the natural environment. studies have shown that other development skills, including Numerous risk variables in the physical environment perceptual and motor abilities that overlap with cognitive have been linked to the severity of accidents between abilities, affect the safety of child-pedestrian interactions child-pedestrians and vehicles. These include road and (Pitcairn & Edlmann, 2000; Schwebel et al., 2012). traffic characteristics, weather and lighting conditions, the However, according to the World Health Organization passage of time, the play area, and the neighborhood (WHO) (Peden, 2004), child-pedestrians of all ages are at (Agran et al., 1996; Davis & Rice, 1994; Dissanayake risk of being involved in a road traffic collision. WHO also et al., 2009; Grechkin et al., 2013; Hobday & Knight, highlighted that 38% of children under 19 years old are 2010; Koopmans et al., 2015; Leden et al., 2006; Steinbach injured or killed as pedestrians on the roads each year et al., 2010). For example, Koopmans et al. (2015) exam- (Toroyan, 2015). In Malaysia, the statistics of child-pedes- ined pedestrian collisions in Illinois, US, using a multi- trian fatalities in Malaysia are particularly worrying due to a variate logistic regression model to estimate injury severity. high rate of fatal road collisions (Mohamed et al., 2011). The The findings showed that risk variables related to physical authors highlighted that 12% of children aged 1 to 18 years settings, such as lighting condition, darkness, midblock old were involved in fatal road collisions from 2007 to 2009. position, traffic management, and road quality, were linked This figure is higher than that of high-income countries, with injury severity in children. Additionally, the authors which only have 5% to 10% of fatalities (Sminkey, 2008). In investigated the interplay of two factors: intersection and particular, a recent study by Darus et al. (2018) found a total traffic control. The results indicate that being located in the of 2,243 child-pedestrian (aged 0–18 years) casualties were middle of a block without traffic control has a substantial reported between 2009 and 2012, accounting for about effect on the severity of injuries. 27.9% of the total number of pedestrian casualties. As a Meanwhile, in a case control study done by Stevenson result, this scenario implies that child-pedestrians are among et al. (1995), Australia, the volume of traffic, in combination the most susceptible road users to traffic injury. with the proportion of vehicles exceeding the speed limit, and As stated by the World Health Organization (WHO) the presence of footpaths, significantly increased the likeli- (Sminkey, 2008), a distinguishing aspect of the physical hood of injuries to child-pedestrians. Moreover, in a review environment, particularly the road environment, is the unique study done by Wazana et al. (1997), the volume of traffic (13 contributing variables associated with children. By and large, times), the speed limit (6.0 times), the predominant type of the road environment is constructed by adults for the benefit dwelling (up to 5.5 times), the absence of a play area (5.3 of adults. As a result, this scenario has increased the danger times), the location on the road (4.2 times), the protection of of harm for children. the play area (3.5 times), the proportion of curbside parking (3.4 times), the street mean vehicle speed (3.3 times), the shared driveway (3.2 times), the type of road (2.9 times), and Physical Environment Risk Factors the time of day (up to 2.7 times) were all found to be signifi- The physical environment refers to the elements of the physi- cant in increasing the probability of child-pedestrian injury. cal surroundings that contribute to the occurrence of poten- Thus, the primary objective of this research is to ascertain tial injury or produce events that cause injury (Haddon, the impact of physical and traffic environment characteris- 1980; Runyan, 2003). The physical environment can be clas- tics on the severity of accidents involving child-pedestrians sified as natural or man-made (Peek-Asa & Zwerling, 2003). aged 18 years or younger in Malaysia. The models are built In the Haddon Matrix model developed by Willam Haddon specifically utilizing multinomial logistic analysis (MNL), Jr., the physical environment is one of the important risk fac- which include critical information about the physical envi- tors within the conceptual framework in all types of injury ronment. The models produced may be used to inform inter- control (Haddon, 1980; Peek-Asa & Zwerling, 2003; vention and environmental change efforts aimed at reducing Runyan, 2003). While, Peek-Asa and Zwerling also pointed injury severity. out that one of the best injury prevention strategies is by changing the physical environment. Generally, by under- Materials and Methods standing the association between the natural/man-made envi- ronment and injury risk, it may suggest significant avenues Data Source and Method of Analysis for injury intervention (Peek-Asa & Zwerling, 2003). In relation to road injuries, the physical characteristics of The dataset was extracted from the Royal Malaysia Police the roadway, such as road conditions, the number of traffic (RMP) by the Malaysian Institute of Road Safety Research lanes, the road’s location, the speed limit, and the volume of (MIROS). The dataset consists of physical environmental Darus et al. 3 Table 1. Terms Defined by the Royal Malaysian Police (RMP) for Road Collisions (Royal Malaysia Police, 2014). Term Definition Pedestrian Any person who is not in a vehicle but occupies a portion of the road; including road construction workers, a person pushing a broken-down vehicle, etc. Fatal road accident A road accident in which one or more people were killed within 30 days of the event’s date. Serious injury road accident A road accident in which at least one person sustained serious injury but none was killed. Slightly/minor injury road accident A road accident in which one or more people were injured, but none were killed or seriously injured. Non-injury road accident A road accident in which no person was killed or injured. characteristics which provide information on the roadway International Comparison of Road Accident characteristics, weather and traffic conditions, and the colli- Database sion time and location. An international comparison database may lead to an increase A total of 2,518 vehicle collisions involving child-pedes- in our understanding of road safety and subsequent counter- trians (events based) were analyzed after excluding about measures in different countries. For this purpose, Table 2 was 4.6% of the total records due to discrepancies such as over- adapted from Montella et al. (2013) and Casado-Sanz et al. lapping report numbers, incomplete records, etc. The analysis (2019) to describe the international comparison between focused on collisions involving child-pedestrians (≤18 years) national databases of selected countries (US, New Zealand, aged 18 years and below for a 6-year period from January Australia, Spain databases including the requirement of EU 2009 to December 2014 in 14 states of Malaysia. Directive). Twenty eight variables were considered for the In this study, the multinomial logit model (MNL) and the comparison. It is noted that in Malaysia, police reports do not associated relative risk ratio (RRR) are obtained and esti- link to the hospital data. Also, specific information such as mated using the Statistical Package for the Social Sciences road segment gradient, average daily traffic (ADT), curve (SPSS) version 23. The Small-Hsiao test of IIA assumption radius and length are not included in the database. is estimated using the Stata 17 software package. Theory of Multinomial Logit Model Malaysia Accident Database The severity levels of the collision were adopted as the Data source dependent variable. These severity levels were defined and The source of data was recorded by the Royal Malaysia recorded by the Royal Malaysian Police (RMP) as shown in Police (RMP) for all types of road accidents. RMP started Table 1. A total of three levels of severity are considered, using the revised accident reporting forms known as Form which are slight/minor injury collision, serious injury colli- POL 27 nationwide on 1st January 1992 (Saidon & Baguley, sion and fatal injury collision. In this study, a multinomial 1994). Overall, the form includes 91 variables, as reported logit model was used to increase flexibility in the model by Ahmed et al. (2020). Basically, the form “POL 27” is specification (Rifaat et al., 2011; Tay et al., 2011). The model divided into eight sections which include reference details of was developed as described in previous studies (Çelik & the report/time of occurrence, carriageway details, environ- Oktay, 2014; Rifaat et al., 2011; Shankar et al., 1996; Tay ment, location, details of the vehicle, details of the driver, et al., 2011; Ulfarsson & Mannering, 2004). details of passengers and pedestrians, and also sketch dia- The probability of a child-pedestrian collision nth with a grams of the accident and location. Then, the data will be severity level i, is as follows: keyed in manually into the Computerized Accident Recording System (CARS) (Hashim & Rahim, 2009). ′ ′ (1) PP =≥ UU ,, ∀i Ii ≠ I () ni ni ni Definitions. In Malaysia, severity of collision is divided where I denotes a set of possible mutually exclusive severity into four levels; fatal, serious, slight, and non-injury. There levels of collisions. are no pedestrian-vehicle accidents recorded as non-injury The U can be expressed as follows: collisions in this data set. As a result, only three severity ni levels are evaluated. The current research makes use of the Ux =+ βε (2) ni in ni definitions set out in Table 1. In contrast to worldwide sta- where β denotes a vector of coefficients, while x denotes a tistics, the majority of nations utilized the same definition vector of explanatory variables, and ε denotes an unobserv- of a fatal traffic accident as Malaysia, namely (dying within ni able random error. ε is assumed to have a generalized 30 days after a crash occurs). However, there are no estab- ni extreme value distribution. From the assumption, the MNL lished definitions of (severe) injuries, and they may vary can be expressed as follows: across nations (OECD, 2010). 4 Table 2. International Comparison in the Road Accident Database. Casado-Sanz et al. (2019). a b Variable EU Directive US MMUCC Australia New Zealand Spain Malaysia Crash location Precise as possible Road name, GPS coordinates Road name, reference Road name, GPS Road name, km Road name, km, location point, distance, direction coordinates GPS coordinates Crash narrative No No Yes Yes Yes Yes Crash sketch No No Yes, access restricted Yes Yes Yes Crash type Yes Recorded in the traffic units Yes Yes Yes Yes section Collision type Yes 8 descriptors Yes Yes 33 descriptors Yes Contributing No Environmental circumstances Yes Yes Yes Yes circumstances Weather conditions Yes 10 descriptors Yes 5 descriptors 9 descriptors Yes Light conditions Yes 7 descriptors Yes 7 descriptors Yes Yes Reported crashes Not specified All severities All injury severities All severities All severities All severities Definition of non-fatal Severe and non- A: Suspected serious injury Injured, admitted to Serious: Requiring Hospitalized, injured Severe severe injuries B: Suspected minor injury hospital Injured, required medical treatment Non-hospitalized, Slight C: Possible injury medical treatment Minor: Other injuries injured Damage Fatalities Within 30 days Within 30 days Within 30 days Within 30 days Within 30 days Within 30 days Link with hospital data No No In Western Australia No Yes No Speed limit Yes Yes Yes Yes Yes Yes Surface conditions Yes 10 descriptors Yes 3 descriptors 9 descriptors 3 descriptors Road curve No Yes Yes 4 descriptors 5 descriptors Yes Road segment gradient No Yes No No No No Age Yes Date of birth Yes Yes Yes Yes Gender Yes Yes Yes Yes Yes Yes Nationality Yes No Foreign drivers identified Foreign drivers identified Yes Yes Injury status No 5 descriptors 4 descriptors Yes 5 descriptors Yes Driver action No 19 descriptors In crash narrative In crash narrative 23 descriptors Yes Pedestrian action No 11 descriptors In crash narrative In crash narrative 11 descriptors Yes Violation codes No Yes Yes Yes No No Safety equipment Yes Yes Yes Yes Yes Yes Seating position No Yes Yes Yes Yes Yes ADT No Yes No Yes No No Curve radius No Yes No Yes No No Length No Yes No Yes No No Source. Table adapted from Montella et al. (2013) and Casado-Sanz et al. (2019). a b Directive 2008/96/EC of the European Parliament and of the Council of 19 November 2008 on road infrastructure safety management, Minimum Uniform Crash Criteria Model (MMUCC). Guideline c d Model minimum uniform crash criteria. National Highway Traffic Safety Administration (NHTSA), only pedestrian and cyclist ages in coded crash listing. Other ages in police crash reports, speed limit recorded since 2015, average daily traffic. Darus et al. 5 ββiXn independent variables as all the VIF values were less than 2. P = ββiXn ni (3) Second, the Small-Hsiao test of the IIA assumption was ∑∀II ′∈ examined (Table 5). The results reject the null hypothesis. It where β is a vector of coefficient that can be estimated for can be concluded that the IIA assumption has not been vio- severity level i using standard maximum likelihood methods. lated. In the next step, multinomial logistic regression analy- As the explanatory variables do not vary across injuries, the sis was performed using a stepwise backward elimination I−1 log-odd ratios of the model become: method to estimate the model. Table 6 adopted a design by Tay et al. (2011) and slight injury collision was selected as ni the reference category based on previous studies (Tay et al., ln == ββ xx −= ββ − xi ,, =… 11 …= I () in In iI n (4) ni 2011). As demonstrated in Table 6, the likelihood ratio chi- square of 302.573 with a small p-value of <.0001 demon- Only the difference in coefficient is identifiable, therefore, strates that the model fits the data significantly. The results the coefficient is identifiable up to an additive constant. The demonstrate that 11 explanatory variables had significant coefficients of one outcome (the base case) are set to zero to relationships between specific road and traffic environment resolve inter determinacy (Carson & Mannering, 2001). The characteristics and the injury severity level of child-pedes- slight injury collisions are used as a base case following trian collisions. The variables with a significant level at (Amoh-Gyimah et al., 2017; Tay et al., 2011). α = .1, α = .05, and α = .01 were included (Pour-Rouholamin As the MNL is a nonlinear model, the estimated coeffi- & Zhou, 2016; Rifaat et al., 2011; Tay et al., 2008). It should cients of the explanatory variables (independent variables) be noted that some variables were retained, although only do not represent their effect on the dependent variables. So, one category in the same factor demonstrated a significant the relative risk ratio (RRR) was computed relative to the value (Kockelman & Kweon, 2002; Tay et al., 2011). The base category to show the effect of significant risk factors. effects of each significant variable were interpreted using the From equation (4), the relative probability of an injury acci- estimated coefficients and relative risk ratios (RRR). dent (i = 2) to the base case (i = 1) is: With regard to geometry segment characteristics, cross Pr i = 2 2 () intersections and T/Y intersections lead to severe and fatal xβ() (5) injury collisions. Serious injury collisions are 1.8 times Pr i = 1 () more likely to occur at the cross intersection (p = .084, RRR = 1.819) than slight injury collisions. On the other The relative risk ratio (RRR) for binary variables is written hand, child-pedestrians who were hit at the T/Y intersection as follows: are 1.7 times more likely to result in fatal injury collisions (6) RRR = e (p = .042, RRR = 1.749). Generally, the RRR refers to an increase if the RRR is more The estimate findings showed that accidents involving seri- than 1 (RRR > 1) or a decrease if the RRR is less than 1 ous injuries occurred at the cross intersection (four-legged (RRR < 1) in the risk of a specific severity level of collision intersection) and the T/Y intersection (three-legged intersec- compared to the base case (Çelik & Oktay, 2014; Rifaat tion). In a study conducted in New York by Ukkusuri et al. et al., 2011). SPSS version 23 was used to estimate the MNL (2012), it was discovered that four-legged and five-legged and the RRR. intersections were linked with an increase in collision fre- quency. This is consistent with the fact that a larger number of legged intersections results in a greater number of conflict Results and Discussion sites, implying a greater probability of collisions at intersec- From the total of 2,518 vehicle collisions involving child- tions (Lee et al., 2016). Ewing and Dumbaugh (2009) also pedestrians, 14.5% were classified as fatal injury collisions, noted that vehicle-vehicle and vehicle-pedestrian accidents 30.1% were classified as serious injury collisions, and 55.4% occur mostly near intersections. Additionally, several studies were classified as slight/minor injury collisions. Table 3 have demonstrated that the density of road intersections may presents the distribution of child-pedestrian collisions based also have a traffic impact on child-pedestrians. For example, on 16 risk factors and 3 injury severity levels. previous scholars (Dissanayake et al., 2009; Steinbach et al., 2010) revealed that the higher density of road intersections was associated with higher child-pedestrian casualties. In a Multinomial Logit Model recent study conducted in Ulsan, Korea, Lee et al. (2016) dis- The estimation of the multinomial logit model is displayed in covered that the population density, number of marked cross- Table 6. First, multicollinearity was tested during the vari- walks, main road width, and number of building entrances at able selection process. There was no multicollinearity issue the intersection were associated with the occurrence of child- in the current study (Table 4). Multicollinearity among the pedestrian collisions at intersections. independent variables was checked using the VIF test. It was The severity of injury collisions is significantly associ- found that there was no multicollinearity issue among the ated with the types of road surface materials on collision 6 SAGE Open Table 3. Distribution (%) of Collision Severity and Contributing Factors. Injury severity levels of collisions (%) Variables Slight/minor Serious Fatal Total Road geometry Bend 5.4 5.9 6.0 5.6 Roundabout 0.2 0.1 0.3 0.2 Cross junction 1.4 2.2 0.3 1.5 T/Y junction 4.2 3.4 6.0 4.2 Staggered junction 0.1 0.1 0.3 0.2 Interchanges 0 0 0.3 0 Straight 88.7 88.1 86.9 88.3 Road surface material Crusher run 2.9 1.3 3.6 2.5 Brick 6.1 5.4 6.3 5.9 Concrete 0.4 0.7 1.1 0.6 Earth 0.5 0.7 3.0 0.9 Bitumen 90.1 91.9 86.1 90.1 Road surface defect Surface defect 2.2 2.0 3.3 2.3 Good condition 97.8 98.0 96.7 97.7 Shoulder type Paved 41.9 38.2 41.3 40.7 Unpaved 58.1 61.8 58.7 59.3 Road surface condition Wet 2.7 4.8 4.9 3.7 Others (oily, sandy, flood, construction works) 1.1 0.9 1.4 1.1 Dry 96.2 94.3 93.7 95.3 Traffic system One-way traffic (single carriageway) 15.9 13.3 15.6 15.1 Two-way traffic (Three lanes divided with bidirectional flow) 1.1 1.8 3.0 1.6 Dual carriageway (Multilane divided with bidirectional flow) 1.3 2.6 5.7 2.3 Two-way traffic (undivided with bidirectional flow) 81.6 82.2 75.7 80.9 Road marking category Double lane line (overtaking not permissible) 9.7 9.6 10.1 9.7 One-way lane line marking 4.9 2.8 3.8 4.1 Divider line marking 5.2 5.0 8.7 5.7 U-turn 0.1 0 0.3 0.1 No marking 24.4 20.7 15.6 22.0 Single lane line (overtaking permissible) 55.8 61.8 61.5 58.4 Control type Police 0.8 0.3 2.2 0.8 Other agencies 0.9 1.1 1.4 1.0 Traffic light 1.8 0.8 0.5 1.3 Pedestrian crossing 1.6 1.6 1.4 1.5 Pedestrian crossing with traffic light 0.4 0.7 0.5 0.5 Level crossing 0.0 0.1 0 0 Yellow line 2.9 1.8 3.3 2.7 Yellow box 1.9 0.9 0.5 1.4 No control 89.6 92.7 90.2 90.6 Speed limit 50 km/h 29.7 24.3 17.5 26.3 70 km/h 17.2 16.8 21.9 17.8 80 km/h 3.8 6.5 9.6 5.4 (continued) Darus et al. 7 Table 3. (continued) Injury severity levels of collisions (%) Variables Slight/minor Serious Fatal Total 90 km/h 7.9 14.3 15.3 10.9 110 km/h 0.5 0.1 0 0.3 Others 40.9 38.0 35.8 39.3 Time collision 0700 to 0959 8.3 8.8 12.1 9.0 1600 to 1859 31.3 32.8 23.3 30.6 1900 to 2459 14.3 16.8 18.8 15.7 0100 to 0659 3.2 5.1 7.0 4.3 1000 to 1559 42.9 36.6 38.8 40.4 Day of week Weekend 25.7 24.0 27.0 25.4 Weekday 74.3 76.0 73.0 74.6 Weather condition Windy 0.4 0.1 0.3 0.3 Foggy 3.9 5.7 6.8 4.9 Rain 2.4 3.3 3.3 2.8 Clear 93.3 90.9 89.6 92.0 Lighting condition Dawn/Dusk 6.3 7.5 7.4 6.8 Dark with street light 8.2 8.3 14.2 9.1 Dark without street light 3.4 5.7 7.1 4.6 Day 82.1 78.5 71.3 79.4 Type of location City 6.2 4.1 3.0 5.1 Urban 12.0 9.4 15.6 11.8 Built-up area 17.6 15.7 12.8 16.4 Rural 64.2 70.8 68.6 66.8 Land use Office 0.019 0.949 1.019 Shopping −0.148 0.525 0.863 Industrial/construction −0.464 0.303 0.629 School 16.1 14.1 6.8 14.2 Others 47.5 56.3 58.2 51.7 Residential 23.7 20.2 25.4 22.9 Hit and run Yes 9.5 8.2 9.0 9.1 No 90.5 91.8 91.0 90.9 Note. Location type: Metropolitan/city-Population more than 75,000 (Elsayed, 2012). Urban area-Gazetted areas with their adjoining built-up areas, which had a combined population of 10,000 or more at the time of the Census 2010 or the special development area that can be identified, which at least had a population of 10,000 with at least 60% of population (aged 15 years and above) were involved in non-agricultural activities (Department of Statistics Malaysia, 2020), Built-up areas were contiguous to a gazetted area and had at least 60% of their population (aged 15 years and above) engaged in non- agricultural activities (Department of Statistics Malaysia, 2020), Rural area is an area with a population less than 10,000 people, having agriculture and natural resources either clustered, linear, or scattered (Hamid & Toyong, 2014). severity and three types of road surface materials. The esti- earth surfaces (RRR = 9.697), respectively. The findings mation results indicated that serious injury collisions are less showed that the presence of concrete and the ground surface likely to occur on crusher run surfaces (p = .041, RRR = 0.471) enhanced the likelihood of sustaining a fatal injury. The in comparison to bitumen. Meanwhile, child-pedestrians crusher run surfaces may serve as a warning to drivers to who were hit by vehicles on concrete (p = .027) and earth slow down, thus reducing the effect of collisions on young- surfaces (p < .001) significantly resulted in fatal injury colli- sters on the road. While the flat surfaces of concrete and dirt sions. The fatal collisions are nearly 4.4 times (RRR = 4.435) result in greater impact collisions. These instances may be and 9.7 times more likely to occur on concrete surfaces and associated with the function of surface roughness. 8 SAGE Open Table 4. Results of the Multicollinearity Test. Standardized coefficients Collinearity statistics Beta T stat Tolerance VIF (Constant) 6.219 Road geometry –.007 –.336 .984 1.017 Road surface material .008 .402 .980 1.020 Road surface defect –.018 –.893 .970 1.031 Shoulder type .007 .312 .897 1.114 Road surface condition .057 2.686 .864 1.158 Traffic system .051 2.522 .968 1.033 Control type –.013 –.637 .961 1.041 Day of week .003 .164 .992 1.008 Weather condition –.048 –2.256 .870 1.149 Lighting condition .052 2.379 .818 1.222 Type of location –.062 –2.945 .892 1.121 Land use .015 .731 .931 1.074 Hit and run .011 .561 .990 1.010 Table 5. Small-Hsiao Test of IIA Assumption. 2 2 lnL (full) lnL(omit) Chi df p > chi Evidence Slight –448.2 –310.785 275.0 54 0.000 Against Ho Serious –1,056.7 –388.9 1,335.7 54 0.000 Against Ho Fatal –1,527.4 –667.4 1,720.2 54 0.000 Against Ho Note. A significant test is evidence against Ho. Additionally, prior research established that the roughness related to all age groups. This could be due to wet road sur- level was strongly associated with the frequency of vehicle faces affecting tire-pavement skid resistance, which is one of collisions and the degree of injuries (Anastasopoulos & the important factors in controlling vehicle direction, speed Mannering, 2009, 2011). Nevertheless, no literature was and short braking distances (Kokkalis & Panagouli, 1998), found on road surface conditions related to the child-pedes- and might have a higher collision impact on child-pedestri- trian severity collision. ans on roads. Other studies confirmed that friction between Overall, the association between severity levels of colli- tires and wet roads had a significant effect on vehicle colli- sions and road surface conditions was found to be statisti- sions (Najafi et al., 2017), as well as ineffective braking sys- cally significant. The results of the current study indicate that tems, causing vehicles to have less control (Graham & the risk of serious injury collisions generally increases on Glaister, 2003). wet road surfaces. The occurrence of serious injury colli- The impact of traffic system categories was also exam- sions is 1.6 times more likely to occur in wet surface condi- ined in this study. It should be noted that the classification of tions (p = .044, RRR = 1.645) compared with slight injury the traffic system category level is adopted from “Form collisions. Similarly, previous studies have reported the POL27” designed by the Royal Malaysian Police. Every cat- effects of wet road surfaces on child-pedestrian collisions. egory level varies with regard to the number of lanes, the For example, Koopmans et al. (2015) demonstrated that wet type of median (divided or undivided), or the type of flow surface conditions were significantly associated with child- (bidirectional or unidirectional). The findings indicated that pedestrian (aged ≤19 years old) fatalities. Meanwhile, a collisions occurring on three lanes (two-way divided) and study done in South Africa by Hobday and Knight (2010) dual carriageways (divided) are positively significant with showed that the highest percentage of fatal injury collisions both serious (three lanes, p = .07 and dual carriageways, involving child-pedestrians (<15 years old) occurred on p = .014) and fatal injury collisions (three lanes, p = .048 and freeways involving buses and trucks in wet surface condi- dual carriageways, p = .001) with reference to the two-way tions. In addition, the relative risk of far-side accidents had a (undivided) system. Serious injury collisions are two times significant effect when the road surface was wet, as reported more likely to occur on three lanes (RRR = 2.046) and 2.3 by Dunbar (2012). This exploratory variable was tested for a times (RRR = 2.327) on dual carriageways. Moreover, child- specific age group and it was found that it was significantly pedestrians are 2.5 times and 4.1 times more likely to sustain 9 Table 6. Multinomial Logit Model of Child-Pedestrian Collisions Severity. Number of observations: 2,518 Log-likelihood at zero: 3,234.843 Log-likelihood at convergence: 3,537.416 Chi-square: 302.573 p-value < .001 Base case: Slight crash injuries Serious crash injuries Fatal crash injuries Category Explanatory variable Reference B Sig Relative risk ratio B Sig Relative risk ratio Intercept –0.878 0.000 –1.647 0.001 Geometry Road geometry Straight Bend 0.022 0.911 1.023 0.018 0.947 1.018 Roundabout –0.543 0.643 0.581 0.475 0.696 1.609 Cross junction 0.598 0.084* 1.819 –1.404 0.176 0.246 T/Y junction –0.128 0.604 0.880 0.559 0.042** 1.749 Staggered junction 0.150 0.904 1.162 0.915 0.470 2.498 Interchanges 1.499 – 4.475 23.206 0.998 1.1E Road surface material Bitumen Crusher run –0.753 0.044** 0.476 0.258 0.472 1.294 Brick –0.104 0.609 0.902 0.030 0.906 1.031 Concrete 0.704 0.253 2.022 1.502 0.026** 4.489 Earth 0.382 0.527 1.466 2.214 0.000*** 9.155 Road surface condition Dry Wet 0.498 0.044** 1.645 0.415 0.183 1.514 Others –0.051 0.916 0.950 –0.160 0.787 0.852 Traffic Traffic system Two-way traffic (undivided with bidirectional flow) One-way traffic (single carriageway) 0.078 0.603 1.081 0.225 0.243 1.253 Two-way traffic (Three lanes divided 0.716 0.070* 2.046 0.900 0.048** 2.461 with bidirectional flow) Dual carriageway (Multilane divided 0.845 0.014** 2.327 1.414 0.001*** 4.112 with bidirectional flow) Road marking category Single lane line (overtaking permissible) Double lane line (overtaking not 0.023 0.891 1.024 0.261 0.239 1.299 permissible) One-way lane line marking –0.506 0.070* 0.603 –0.428 0.219 0.652 Divider line marking –0.226 0.304 0.798 0.084 0.742 1.087 –7 U-turn –14.911 0.995 3.3 × 10 1.080 0.512 2.945 (continued) 10 Table 6. (continued) Base case: Slight crash injuries Serious crash injuries Fatal crash injuries Category Explanatory variable Reference B Sig Relative risk ratio B Sig Relative risk ratio No marking –0.135 0.291 0.874 –0.559 0.004*** 0.572 Control type No control Police –1.148 0.140 0.317 0.858 0.096* 2.359 Other agencies 0.336 0.475 1.399 0.338 0.576 1.402 Traffic light –0.858 0.073* .424 –2.220 0.035** 0.109 Pedestrian crossing 0.131 0.731 1.140 0.099 0.860 1.104 Pedestrian crossing with traffic light 0.651 0.295 1.918 0.757 0.395 2.132 Level crossing –15.916 0.995 8.3 × 10 –0.441 1.000 0.644 Yellow line –0.420 0.198 0.657 0.470 0.198 1.600 Yellow box –0.663 0.130 0.515 –1.213 0.110 0.297 Speed limit 50 km/h 70 km/h 0.052 0.727 1.055 0.508 0.011** 1.662 80 km/h 0.571 0.012** 1.754 1.175 0.000*** 3.238 90 km/h 0.632 0.000*** 1.844 0.851 0.000*** 2.342 –7 110 km/h –1.222 0.261 0.292 –13.910 0.990 9.1x10 Others 0.063 0.599 1.060 0.366 0.039** 1.442 Temporal trend Time collision 1000 to 1559 0700 to 0959 0.190 0.284 1.209 0.360 0.097* 1.433 0000 to 0659 0.191 0.093* 1.210 0.639 0.021** 1.895 1600 to 1859 0.299 0.119 1.348 –0.310 0.058* 0.734 1900 to 2459 0.490 0.035** 1.632 –0.285 0.268 0.752 Environmental Lighting condition Day Dawn/Dusk 0.041 0.835 1.041 0.180 0.493 1.198 Dark with street light –0.077 0.721 0.926 0.771 0.003*** 2.162 Dark without street light 0.287 0.275 1.332 0.883 0.006*** 2.418 Land use Type of location Rural characteristic City –0.486 0.041** 0.615 –0.915 0.010*** 0.400 Urban –0.250 0.130 0.779 0.154 0.439 1.166 Built-up area –0.063 0.636 0.938 –0.227 0.231 0.797 Land use Residential Office 0.019 0.949 1.019 –0.418 0.298 0.658 Shopping –0.148 0.525 0.863 –0.871 0.014** 0.418 Industrial/construction –0.464 0.303 0.629 0.554 0.179 1.740 Bridge 0.212 0.668 1.236 –1.559 0.148 0.201 School 0.039 0.817 1.040 –1.135 0.000*** 0.321 Others 0.173 0.116 1.188 –0.102 0.521 0.903 *Significant at the 90% confidence interval, **significant at the 95% confidence interval, and ***significant at the 99% confidence interval. Slight injury crash used as a base case or reference. Darus et al. 11 fatal injuries if they are hit on three lanes (RRR = 2.461) and previous studies (Çelik & Oktay, 2014; Lee & Abdel-Aty, dual carriageways (RRR = 4.112). The results indicate that 2005; Rifaat et al., 2011). In the presence of traffic signals, child-pedestrians are at a greater risk of fatality and serious drivers might be more cautious and tend to slow down, injuries in traffic system categories with multilane roadways thereby resulting in a lower impact of injury collisions (Lee (more than two lanes), divided medians, and bidirectional & Abdel-Aty, 2005; Rifaat et al., 2011). Nevertheless, fatal traffic flow (two-way). This result is in agreement with a injury collisions (p = .086, RRR = 2.430) are 2.4 times more study conducted in Hong Kong (Sze & Wong, 2007) and likely to occur at a location controlled by the police. No ref- Riyadh (Al-Ghamdi, 2002). Besides, Wong et al. (2007) erence was found related to the location controlled by the found that two-way and multi/dual carriageways were sig- police. nificantly related to an increased probability of severe injury Additionally, the effect of various kinds of speed limit levels. While, Al-Ghamdi (2002), reported that two-way indications was investigated, with roads having a speed limit divided roadways (two-way with median) are the most fre- indication of 70 km/h (p = .007, RRR = 1.726) being substan- quent locations for pedestrian-vehicle collisions in Riyadh. tially linked with fatal injuries. Additionally, highways with Nonetheless, several studies examined each characteristic of speed limit signs of 80 and 90 km/h were strongly linked the traffic system associated with severity level separately. with severe and fatal injury accidents. Additionally, acci- For example, several studies (Agran et al., 1996; Mueller dents involving speeds of 80 km/h (p = .001, RRR = 3.417) et al., 1990) found that roadways with more than two travel and 90 km/h (p = .001, RRR = 2.347) are 3.4 times and 2.3 lanes and wider lanes are related to higher risks of child- times more likely to result in fatal injuries, respectively. This pedestrian collisions. The results indicate that wider lanes result indicates that highways having posted speed limits of may be correlated with higher traffic speed and lead to higher 70 km/h more are linked with a higher severity level. occurrence rates and risks of child-pedestrian collisions Similarly, prior research has shown that locations with higher (Abdel-Aty et al., 2007; Al-Ghamdi, 2002; Ewing & speed limit signs are associated with higher impact speeds Dumbaugh, 2009). Nevertheless, Abdel-Aty et al. (2007), and a larger probability of fatal collisions (Davis, 2001). revealed that the higher collision involvement among school- Additionally, Davis (2001) created a discrete outcome model aged children in Florida increases as the number of lanes that links the degree of an injured pedestrian’s injury to the increases on divided roads. This study indicated that a multi- hitting vehicle’s speed. Additionally, it was shown that child- lane roadway with a median presence influences the risk of pedestrians aged 0 to 14 years old are more likely to have collisions. With regard to the type of flow, Al-Ghamdi (2002) severe injuries when collision speeds exceed 40 km/h. found that high levels of injuries were associated with two- Additionally, if they are struck at a speed greater than way roadways with a median. Therefore, the risk of injury 75 km/h, fatal injuries are quite probable. collisions may be caused by any characteristics of the traffic With respect to the time of collisions, the findings revealed system, which are the number of lanes, the type of median that collisions occurring from 000 to 0659 hours (midnight to (divided or undivided), and the type of flow (bidirectional or early morning) were significantly related to serious (=0.093, unidirectional), or any combination of the aforementioned RRR = 1.210) and fatal injury collisions (p = .021, characteristics. RRR = 1.895). Also, collisions are likely to be fatal (p = .097, As for the road marking category, serious (p = .07, RRR = 1.433) from 0700 to 0959 hours (morning). While, RRR = 0.603) collision injuries were found to be less likely serious injury collisions are likely to occur from 1900 to to occur on roads with a one-way line. Also, fatal (p = .004, 2459 hours (night). Nevertheless, collisions are less likely to RRR = 0.572) collision injuries are less likely to occur on cause fatal injury collisions from 1600 to 1859 hours (late roads with no marking lines. As such, serious injury colli- afternoon). This finding implies that child-pedestrians in sions are 1.7 times (1/0.603) and fatal injury collisions are Malaysia experienced more severe injury collisions during almost 1.7 times (1/0.572) more likely to occur on roads with the early morning, morning, and night time. This may be a single line compared to roads with a one-way line and no related to peak traffic periods and school opening and clos- marking line, respectively. This result might be due to the ing times, as observed in previous studies (Pitt et al., 1990; fact that vehicles are allowed to overtake other vehicles on Yiannakoulias et al., 2002). Nevertheless, fatal collisions roads with a single line, which can lead to higher severity of during midnight might be due to negligence factors by other collisions. For example, Leden et al. (2006) found that child- road users. Moreover, fatigue, tiredness, stress, speeding, pedestrians in Finland were frequently involved in collisions and lack of visibility at night may be attributed to driver fac- involving overtaking vehicles at mid-block. With regard to tors (Amoh-Gyimah et al., 2017). traffic control devices, the results of the current study dem- The natural lighting conditions were found to be signifi- onstrate that collisions at traffic lights were found to be less cant with fatal injury. Dark conditions with street lights likely to result in serious (p = .073, RRR = 0.424) and fatal (p = .003, RRR = 2.162) are positively associated with fatal injury collisions (p = .035, RRR = 0.109). This finding indi- injury collisions. Additionally, fatal injury collisions are 2.4 cated that the presence of traffic lights reduced the probabil- times more likely to occur in the dark without street light ity of serious and fatal injuries, which was in agreement with (p = .006, RRR = 2.418) compared to slight injury collisions. 12 SAGE Open This result indicates that child-pedestrians are at a higher et al. (2007) found that the presence of a driveway and turn- injury risk in darkness. This result is consistent with previous ing bay decreases the incidence rates and severity of injuries findings (Amoh-Gyimah et al., 2017; Koopmans et al., 2015; in school areas. Nevertheless, this finding contradicts Abdel- Mohamed et al., 2013; Rifaat et al., 2011), which demon- Aty et al. (2007), whose study demonstrated that school chil- strated that darkness may affect the visibility of drivers and dren are at greater risk due to the fact that the middle and pedestrians. Rifaat et al. (2011) highlighted that drivers may high schools in Orange County, Florida, tend to be located struggle to identify hazards and other road users. Additionally, near multi-lane high-speed roads. drivers may be unable to perform appropriate evasive maneu- vers and slow down, resulting in more severe accidents. Conclusion In relation to location characteristics, the results reveal that city areas were less likely to have serious (p = .041, The current study identified several risk factors associated RRR = 0.615) or fatal injury collisions (p = .010, RRR = 0.400) with the increasing severity of vehicle collisions involving compared to rural areas. This result indicates that child- child-pedestrians. The explanatory variables associated with pedestrians in urban areas may be exposed to fewer severe increased probability of fatal injury collisions include: t/y injuries compared to those who live in rural areas. This find- intersection; concrete and earth-road surfaces; two-way traffic ing is similar to a few other studies (Afukaar et al., 2003; and dual carriageways; posted speed limits of 70 to 90 km/h; Doukas et al., 2010; Singh et al., 2016). Evidence indicated time of collision: 0 to 0659 hours (early morning) and 0700 to that lack of road maintenance, poor quality of public trans- 0959 hours (morning); lighting conditions, including dark port, low awareness of traffic rules, higher vehicle speeds, without street lights and dark with street lights; and control fewer separated pedestrian facilities, and limited access to type involving police. Also, cross intersections, posted speed medical emergency services might be contributing factors to limits of 80 to 90 km/h, and time of collision, which are from severe injuries in rural areas (Afukaar et al., 2003; Singh 0 to 0659 hours (early morning) and from 1900 to 2459 hours et al., 2016). This finding shows that the specific carriage- (night), significantly increased serious injury collisions. way features in different locations will impact the road From a traffic engineering perspective, road characteris- accident severity levels. Alternatively, past researchers tics, road designs, and traffic operations play an important (Yiannakoulias et al., 2002) also suggested that the associa- role in providing safe walking conditions for vulnerable road tion between physical environmental characteristics in com- users (Oxley et al., 2018). Fundamentally, child-pedestrians bination with spatial analysis with different analytical should be provided with good and protective facilities based approaches could be adopted for a comprehensive overview on their abilities (Assailly, 1997) and skills. For example, of the influence of accident location on severity levels. there should be special attention given to locations with com- Besides that, the effect of different types of land use char- plex road geometrics and traffic systems, such as t/y intersec- acteristics was examined. From the model estimation, it was tions, cross intersections, and multilane roads with found that slight crash injuries (p = .014, RRR = 0.418) are bidirectional flows, as all these locations tend to be more nearly 2.4 times (1/0.418) more likely to occur in shopping severe. The reduction in child-pedestrian collisions is associ- areas compared to fatal crash injuries. Similar to this finding, ated with lighting conditions and road surface conditions. It Clifton and Kreamer-Fults (2007) found that commercial should be noted that child-pedestrian collisions may be land uses near public schools were associated with higher attributed to drivers’ negligence and irresponsible behavior. pedestrian-vehicular collision severity. Meanwhile Elias and Importantly, dark areas should be installed with adequate Shiftan (2014) and Mohamed et al. (2013), reported that chil- street lighting to increase visibility. Furthermore, road safety dren who live in areas of mixed land use (including commer- educational programs and campaigns should also be targeted cial land use) are significantly exposed to road collisions. at drivers to increase their awareness of pedestrians’ and Clifton and Kreamer-Fults (2007) revealed that the location cyclists’ activities (Desapriya et al., 2011; Tay et al., 2011). characteristics were likely associated with higher levels of Moreover, modifications to the physical environment should pedestrian demand and thus high absolute numbers of be combined with effective training techniques involving crashes. Elias and Shiftan (2014) demonstrated that metered children, especially in school zones, to reduce the selection parking facilities, which are located in commercial areas of poor routes (Schwebel et al., 2012). Notably, future where speeds tend to be lower, have a significant effect on research should examine the human factors concerning chil- reducing fatality risks. Also, it was found that slight crash dren’s behavior and development in order to better under- injuries (p = .000, RRR = 0.321) are nearly 3.1 (1/0.321) stand child-pedestrians’ abilities and limitations in designing times more likely to occur in school areas. This finding is pedestrian facilities. similar to other studies (Clifton & Kreamer-Fults, 2007; Pitt et al., 1990). Pitt et al. (1990) found that the lower incidence Limitation rates and lower severity of injuries in the vicinity of schools might be correlated to the success of road safety education There are a few limitations to this research. To begin, data on programs and other traffic safety measures, while Clifton physical environmental features was recorded separately Darus et al. 13 from data on human aspects (age, gender, behavior, etc.). As Amoh-Gyimah, R., Aidoo, E. N., Akaateba, M. A., & Appiah, S. K. (2017). The effect of natural and built environmen- a result, association between human factors variables and tal characteristics on pedestrian-vehicle crash severity in physical environmental features has not been examined. Ghana. International Journal of Injury Control and Safety Second, the database does not include spatial features such Promotion, 24(4), 459–468. https://doi.org/10.1080/174573 as the dimensions of pedestrian pathways, the layout and 00.2016.1232274 usage of the area between the road and the sidewalk. It is Anastasopoulos, P. C., & Mannering, F. L. (2009). A note on mod- recommended that the analysis of physical environment eling vehicle accident frequencies with random-parameters characteristics in combination with spatial analysis with dif- count models. Accident Analysis and Prevention, 41(1), ferent analytical approach should be further investigated. 153–159. https://doi.org/10.1016/j.aap.2008.10.005 Another drawback of the research was that the database was Anastasopoulos, P. C., & Mannering, F. L. (2011). An empirical only available until 2014. In terms of future work, it may be assessment of fixed and random parameter logit models using prudent to do further analysis of current data in order to crash- and non-crash-specific injury data. Accident Analysis and Prevention, 43(3), 1140–1147. https://doi.org/10.1016/j. enhance the predictability of the MNL models. Despite data aap.2010.12.024 limitations, these statistics offer sufficient information on Assailly, J. P. (1997). Characterization and prevention of child road and environmental factors for preventive measures. pedestrian accidents: An overview. Journal of Applied Developmental Psychology, 18(2), 257–262. https://doi.org Acknowledgments /10.1016/S0193-3973(97)90039-3 The authors wish to thank Malaysian Royal Police (RMP) and Barton, B. K., Ulrich, T., & Lyday, B. (2012). The roles of gender, Malaysian Institute of Road Safety Research (MIROS) for provid- age and cognitive development in children’s pedestrian route ing road accident data for this project and financial support from selection. Child Care Health and Development, 38(2), Ministry of Education, Malaysia. 280–286. https://doi.org/10.1111/j.1365-2214.2010.01202.x Carson, J., & Mannering, F. (2001). The effect of ice warning signs Declaration of Conflicting Interests on ice-accident frequencies and severities. Accident Analysis and Prevention, 33(1), 99–109. https://doi.org/10.1016/S0001- The author(s) declared no potential conflicts of interest with respect 4575(00)00020-8 to the research, authorship, and/or publication of this article Casado-Sanz, N., Guirao, B., Galera, A. L., & Attard, M. (2019). Investigating the risk factors associated with the severity of the Funding pedestrians injured on Spanish crosstown roads. Sustainability, The author(s) disclosed receipt of the following financial support 11(19), 1–18. https://doi.org/10.3390/su11195194 for the research, authorship, and/or publication of this article: This Çelik, A. K., & Oktay, E. (2014). A multinomial logit analy- research was funded by Ministry of Education, Malaysia research sis of risk factors influencing road traffic injury severi- Grant no. FRGS/1/2021/TK02/UKM/02/1. ties in the Erzurum and Kars provinces of Turkey. Accident Analysis and Prevention, 72, 66–77. https://doi.org/10.1016/j. aap.2014.06.010 ORCID iD Clifton, K. J., & Kreamer-Fults, K. (2007). An examination of Muhamad Nazri Borhan https://orcid.org/0000-0003-4517-5659 the environmental attributes associated with pedestrian- vehicular crashes near public schools. Accident Analysis References and Prevention, 39, 708–715. https://doi.org/10.1016/j. Abdel-Aty, M., Chundi, S. S., & Lee, C. (2007). Geo-spatial and aap.2006.11.003 log-linear analysis of pedestrian and bicyclist crashes involv- Clifton, K. J., Livi Smith, A. D., & Rodriguez, D. (2007). The ing school-aged children. Journal of Safety Research, 38(5), development and testing of an audit for the pedestrian envi- 571–579. https://doi.org/10.1016/j.jsr.2007.04.006 ronment. Landscape and Urban Planning, 80(1–2), 95–110. Afukaar, F. K., Antwi, P., & Ofosu-Amaah, S. (2003). Pattern of https://doi.org/10.1016/j.landurbplan.2006.06.008 road traffic injuries in Ghana: Implications for control. Injury Darus, N. S., Borhan, M. N., Ishak, S. Z., Ismail, R., & Mohd. Control and Safety Promotion, 10(1–2), 69–76. https://doi. Razali, S. F. (2018). Pattern of child pedestrian collisions and org/10.1076/icsp.10.1.69.14107 injuries in Malaysia. Jurnal Teknologi, 80(4), 1–8. https://doi. Agran, P. F., Winn, D. G., Anderson, C. L., Tran, C., & Del Valle, org/10.11113/jt.v80.11177 C. P. (1996). The role of the physical and traffic environment Davis, G. A. (2001). Relating severity of pedestrian injury to in child pedestrian injuries. Pediatrics, 98(6 Pt 1), 1096–1103. impact speed in vehicle-pedestrian crashes: Simple thresh- Ahmed, A., Farhan, A., Sadullah, M., & Yahya, A. S. (2020). old model. Transportation Research Record: Journal of the The effect of type of attributes on the fill-ability of acci- Transportation Research Board, 1773(1), 108–113. dent reporting forms (ARF). International Journal of Traffic Davis, N. K., & Rice, R. G. (1994). The role of the physical envi- and Transportation Management, 1(2), 11–18. https://doi. ronment in child pedestrian accidents. Journal of Advanced org/10.5383/jttm.01.02.002 Transportation, 28(2), 171–187. Al-Ghamdi, A. S. (2002). Pedestrian-vehicle crashes and analytical Department of Statistics Malaysia. (2020). Definition of urban area techniques for stratified contingency tables. Accident Analysis by Department of Statistic. https://www.dosm.gov.my/v1/ and Prevention, 34(2), 205–214. https://doi.org/10.1016/ index.php?r=column/cone&menu_id=bDA2VkxRSU40STcxd S0001-4575(01)00015-X kZ4OGJ0c1ZVdz09 14 SAGE Open Desapriya, E., Sones, M., Ramanzin, T., Weinstein, S., Scime, G., Kokkalis, A. G., & Panagouli, O. K. (1998). Fractal evaluation & Pike, I. (2011). Injury prevention in child death review: of pavement skid resistance variations. II: Surface wear. Child pedestrian fatalities. Injury Prevention, 17, 4–9. https:// Chaos, Solitons, and Fractals, 9(11), 1891–1899. https://doi. doi.org/10.1136/ip.2010.026914 org/10.1016/S0960-0779(98)00185-4 Dissanayake, D., Aryaija, J., & Wedagama, D. M. (2009). Koopmans, J. M., Friedman, L., Kwon, S., & Sheehan, K. (2015). Modelling the effects of land use and temporal factors on child Urban crash-related child pedestrian injury incidence and char- pedestrian casualties. Accident Analysis and Prevention, 41(5), acteristics associated with injury severity. Accident Analysis 1016–1024. https://doi.org/10.1016/j.aap.2009.06.015 and Prevention, 77, 127–136. https://doi.org/10.1016/j. Doukas, G., Olivier, J., Poulos, R., & Grzebieta, R. (2010). Exploring aap.2015.02.005 differential trends in severe and fatal child pedestrian injury in Leden, L., Gårder, P., & Johansson, C. (2006). Safe pedes- New South Wales, Australia (1997–2006). Accident Analysis trian crossings for children and elderly. Accident Analysis and Prevention, 42(6), 1705–1711. https://doi.org/10.1016/j. and Prevention, 38(2), 289–294. https://doi.org/10.1016/j. aap.2010.04.010 aap.2005.09.012 Dunbar, G. (2012). The relative risk of nearside accidents is high Lee, C., & Abdel-Aty, M. (2005). Comprehensive analysis for the youngest and oldest pedestrians. Accident Analysis of vehicle-pedestrian crashes at intersections in Florida. and Prevention, 45, 517–521. https://doi.org/10.1016/j. Accident Analysis and Prevention, 37, 775–786. https://doi. aap.2011.09.001 org/10.1016/j.aap.2005.03.019 Elias, W., & Shiftan, Y. (2014). Analyzing and modeling risk Lee, G., Park, Y., Kim, J., & Cho, G.-H. (2016). Association between exposure of pedestrian children to involvement in car crashes. intersection characteristics and perceived crash risk among Accident Analysis and Prevention, 62, 397–405. https://doi. school-aged children. Accident Analysis and Prevention, 97, org/10.1016/j.aap.2013.06.035 111–121. https://doi.org/10.1016/j.aap.2016.09.001 Elsayed, I. (2012). Effects of population density and land man- Mohamed, M. G., Saunier, N., Miranda-Moreno, L. F., & Ukkusuri, agement on the intensity of urban heat islands: A case study S. V. (2013). A clustering regression approach: A comprehen- on the city of Kuala Lumpur, Malaysia. In B. M. Alam (Ed.), sive injury severity analysis of pedestrian–vehicle crashes in Application of geographic information systems (pp. 267–282). New York, US and Montreal, Canada. Safety Science, 54, IntechOpen. 27–37. https://doi.org/10.1016/j.ssci.2012.11.001 Ewing, R., & Dumbaugh, E. (2009). The built environ- Mohamed, N., Wong, S. V., Hashim, H. H., & Othman, I. (2011). ment and traffic safety: A review of empirical evidence. An overview of road traffic injuries among children in Malaysia Journal of Planning Literature, 23, 347–367. https://doi. and its implication on road traffic injury prevention strategy org/10.1177/0885412209335553 (MRR 03/2011 Research report). Malaysian Institute of Road Graham, D. J., & Glaister, S. (2003). Spatial variation in road Safety Research. pedestrian casualties: The role of urban scale, density and Montella, A., Andreassen, D., Tarko A. P., & Turner, S. (2013). land-use mix. Urban Studies, 40(8), 1591–1607. https://doi. Crash databases in Australasia, the European Union, and the org/10.1080/0042098032000094441 United States. Transportation Research Record, 2386, 128–136. Grechkin, T. Y., Chihak, B. J., Cremer, J. F., Kearney, J. K., & https://doi.org/10.3141/2386-15 Plumert, J. M. (2013). Perceiving and acting on complex Mueller, B. A., Rivara, F. P., Lii, S.-M., & Weiss, N. S. (1990). affordances: How children and adults bicycle across two Environmental factors and the risk for childhood pedestrian- lanes of opposing traffic. Journal of Experimental Psychology motor vehicle collision occurrence. American Journal of Human Perception & Performance, 39(1), 23–36. https://doi. Epidemiology, 132(3), 550–560. org/10.1037/a0029716 Najafi, S., Flintsch, G. W., & Medina, A. (2017). Linking roadway Haddon, W. (1980). Advances in the epidemiology of injuries crashes and tire–pavement friction: A case study. International as a basis for public policy. Public Health Reports, 95(5), Journal of Pavement Engineering, 18(2), 119–127. https://doi. 411–421. org/10.1080/10298436.2015.1039005 Hamid, L. A., & Toyong, N. M. P. (2014). Rural area, elderly OECD. (2010). Reporting on serious road traffic casualties. http:// people and the house breaking crime. Procedia - Social and www.internationaltransportforum.org/irtadpublic/pdf/Road- Behavioral Sciences, 153, 443–451. https://doi.org/10.1016/j. Casualties-Web.pdf sbspro.2014.10.078 Oxley, J., Fildes, B., Ihsen, E., Charlton, J., & Day, R. (1997). Hashim, H. H., & Rahim, S. A. S. M. (2009, September 16–17). The Differences in traffic judgements between young and old construction of road accident analysis and database system in adult pedestrians. Accident Analysis and Prevention, 29(6), th Malaysia [Conference session]. The 4 IRTAD Conference, 839–847. https://doi.org/10.1016/S0001-4575(97)00053-5 Seoul, Korea. http://www.internationaltransportforum.org/irtad Oxley, J., Jamaludin, A., & Johnson, M. (2012). Analysis of child /Seoul2009.html pedestrian deaths and serious injuries in Malaysia. Journal of Hobday, M., & Knight, S. (2010). Motor vehicle colli- the Australasian College of Road Safety, 23(2), 30–37. sions involving child pedestrians in ethekwini in 2007. Oxley, J., O’Hern, S., Burtt, D., & Rossiter, B. (2018). Falling while Journal of Child Health Care, 14(1), 67–81. https://doi. walking: A hidden contributor to pedestrian injury. Accident org/10.1177/1367493509347059 Analysis and Prevention, 114, 77–82. https://doi.org/10.1016/j. Kockelman, K. M., & Kweon, Y. J. (2002). Driver injury sever- aap.2017.01.010 ity: An application of ordered probit models. Accident Analysis Peden, M. (2004). World report on road traffic injury prevention. and Prevention, 34(3), 313–321. https://doi.org/10.1016/ Public Health, 120(3), 280–280. https://doi.org/10.1016/j. S0001-4575(01)00028-8 puhe.2005.09.003 Darus et al. 15 Peek-Asa, C., & Zwerling, C. (2003). Role of environmental injury rates in London. Health & Place, 16(1), 34–42. https:// interventions in injury control and prevention. Epidemiologic doi.org/10.1016/j.healthplace.2009.08.002 Reviews, 25, 77–89. https://doi.org/10.1093/epirev/mxg006 Stevenson, M. R., Jamrozik, K. D., & Spittle, J. (1995). A case- Pitcairn, T. K., & Edlmann, T. (2000). Individual differences in control study of traffic risk factors and child pedestrian injury. road crossing ability in young children and adults. British International Journal of Epidemiology, 24(5), 957–964. https:// Journal of Psychology, 91(3), 391–410. doi.org/10.1093/ije/24.5.957 Pitt, R., Guyer, B., Hsieh, C. C., & Malek, M. (1990). The sever- Sze, N. N., & Wong, S. C. (2007). Diagnostic analysis of the ity of pedestrian injuries in children: An analysis of the pedes- logistic model for pedestrian injury severity in traffic crashes. trian injury causation study. Accident Analysis and Prevention, Accident Analysis and Prevention, 39(6), 1267–1278. https:// 22(6), 549–559. https://doi.org/10.1016/0001-4575(90)90027-I doi.org/10.1016/j.aap.2007.03.017 Pour-Rouholamin, M., & Zhou, H. (2016). Investigating the risk Tay, R., Choi, J., Kattan, L., & Khan, A. (2011). A multinomial factors associated with pedestrian injury severity in Illinois. logit model of pedestrian–vehicle crash severity. International Journal of Safety Research, 57, 9–17. https://doi.org/10.1016/j. Journal of Sustainable Transportation, 5, 233–249. https://doi. jsr.2016.03.004 org/10.1080/15568318.2010.497547 Rifaat, S. M., Tay, R., & de Barros, A. (2011). Effect of street pat- Tay, R., Rifaat, S. M., & Chin, H. C. (2008). A logistic model tern on the severity of crashes involving vulnerable road users. of the effects of roadway, environmental, vehicle, crash and Accident Analysis and Prevention, 43(1), 276–283. https://doi. driver characteristics on hit-and-run crashes. Accident Analysis org/10.1016/j.aap.2010.08.024 and Prevention, 40(4), 1330–1336. https://doi.org/10.1016/j. Rothman, L., Slater, M., Meaney, C., & Howard, A. (2010). Motor aap.2008.02.003 vehicle and pedestrian collisions: Burden of severe injury on Tiwari, G. (2020). Progress in pedestrian safety research. major versus neighborhood roads. Traffic Injury Prevention, International Journal of Injury Control and Safety 11(1), 43–47. https://doi.org/10.1080/15389580903452340 Promotion, 27(1), 35–43. https://doi.org/10.1080/1745730 Royal Malaysia Police. (2014). Laporan Perangkaan Kemalangan 0.2020.1720255 Jalan Raya Malaysia 2014. Polis DiRaja Malaysia. Toroyan, T. (2015). Global status report on road safety. World Runyan, C. W. (2003). Introduction: Back to the future – revisit- Health Orgainisation, p. 318. http://www.who.int/violence_ ing Haddon’s conceptualization of injury epidemiology and injury_prevention/road_safety_status/2015/en/ prevention. Epidemiologic Reviews, 25, 60–64. https://doi. Ukkusuri, S., Miranda-Moreno, L. F., Ramadurai, G., & Isa- org/10.1093/epirev/mxg005 Tavarez, J. (2012). The role of built environment on pedestrian Saidon, N., & Baguley, C. (1994, March). Road accidents in crash frequency. Safety Science, 50(4), 1141–1151. https://doi. Malaysia [Conference session]. Proceeding seminar on crash org/10.1016/j.ssci.2011.09.012 and injury control, Universiti Malaya, Kuala Lumpur. Ulfarsson, G. F., & Mannering, F. L. (2004). Differences in male Sandels, S. (1970). Young children in traffic. British Journal and female injury severities in sport-utility vehicle, minivan, of Educational Psychology, 40(2), 111–116. https://doi. pickup and passenger car accidents. Accident Analysis and org/10.1111/j.2044-8279.1970.tb02111.x Prevention, 36(2), 135–147. https://doi.org/10.1016/S0001- Schieber, R. A., & Thompson, N. J. (1996). Developmental risk 4575(02)00135-5 factors for childhood pedestrian injuries. Injury Prevention, Vinje, M. P. (1981). Children as pedestrians: Abilities and limi- 2(3), 228–236. https://doi.org/10.1136/ip.2.3.228 tations. Accident Analysis and Prevention, 13(3), 225–240. Schwebel, D. C., Davis, A. L., & O’Neal, E. E. (2012). Child pedes- https://doi.org/10.1016/0001-4575(81)90006-3 trian injury: A review of behavioral risks and preventive strat- Wazana, A., Krueger, P., Raina, P., & Chambers, L. (1997). A egies. American Journal of Lifestyle Medicine, 6, 292–302. review of risk factors for child pedestrian injuries: Are they https://doi.org/10.1177/0885066611404876 modifiable? Injury Prevention, 3(4), 295–304. https://doi. Shankar, V., Mannering, F., & Barfield, W. (1996). Statistical anal- org/10.1136/ip.4.3.243-b ysis of accident severity on rural freeways. Accident Analysis Whitebread, D., & Neilson, K. (2000). The contribution of visual and Prevention, 28(3), 391–401. https://doi.org/10.1016/0001- search strategies to the development of pedestrian skills by 4-11 4575(96)00009-7 year-old children. British Journal of Educational Psychology, Singh, D., Singh, S. P., Kumaran, M., & Goel, S. (2016). 70(4), 539–557. https://doi.org/10.1348/000709900158290 Epidemiology of road traffic accident deaths in children in Wong, S. C., Sze, N. N., & Li, Y. C. (2007). Contributory factors Chandigarh zone of North West India. Egyptian Journal of to traffic crashes at signalized intersections in Hong Kong. Forensic Sciences, 6, 255–260. https://doi.org/10.1016/j. Accident Analysis and Prevention, 39, 1107–1113. https://doi. ejfs.2015.01.008 org/10.1016/j.aap.2007.02.009 Sminkey, L. (2008). World report on child injury prevention. Injury Yao, J. F., Yang, J. K., & Fredriksson, R. (2006). Reconstruction Prevention, 14(1), 69. https://doi.org/10.1136/ip.2007.018143 of head-to-hood impact in an automobile-to-child-pedestrian Solagberu, B. A., Osuoji, R. I., Ibrahim, N. A., Oludara, M. A., collision. International Journal of Crashworthiness, 11(4), Balogun, R. A., Ajani, A. O., Idowu, O. E., Mustafa, I. A., 387–395. https://doi.org/10.1533/ijcr.2005.0118 & Sanni, F. O. (2014). Child pedestrian injury and fatality in Yiannakoulias, N., Smoyer-Tomic, K. E., Hodgson, J., Spady, D. a developing country. Pediatric Surgery International, 30(6), W., Rowe, B. H., & Voaklander, D. C. (2002). The spatial and 625–632. https://doi.org/10.1007/s00383-014-3516-8 temporal dimensions of child pedestrian injury in Edmonton. Steinbach, R., Green, J., Edwards, P., & Grundy, C. (2010). Race Canadian Journal of Public Health, 93(6), 447–451. http:// or place? Explaining ethnic variations in childhood pedestrian www.jstor.org/stable/41993563

Journal

SAGE OpenSAGE

Published: Jan 17, 2022

Keywords: child-pedestrian; road injuries; collisions; severity

There are no references for this article.