• Users Online: 2291
  • Print this page
  • Email this page


 
 
Table of Contents
ORIGINAL ARTICLE
Year : 2018  |  Volume : 7  |  Issue : 4  |  Page : 133-139

Spatial factors related to traffic crashes on pedestrians in all districts of Tehran


1 Occupational Health and Safety Research Center, Hamadan University of Medical Sciences, Hamadan, Iran
2 Safety Promotion and Injury Prevention Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran
3 Department of Epidemiology, School of Public Health, Shahid Beheshti University of Medical Sciences, Tehran, Iran
4 Department of Human Geography/Urban Planning, University of Tehran, Tehran, Iran
5 Social Determinants of Health Research Center, Research Institute for Health Development, Kurdistan University of Medical Sciences, Sanandaj, Iran
6 Department of Epidemiology, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran
7 Crashes Office, Tehran's Traffic Police, Tehran, Iran

Date of Web Publication31-May-2019

Correspondence Address:
Prof. Hamid Soori
Safety Promotion and Injury Prevention Research Center, Shahid Beheshti University of Medical Sciences, Tehran
Iran
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/atr.atr_107_18

Rights and Permissions
  Abstract 

Background: A great proportion of deaths due to traffic crashes occur for pedestrians, both in developing and developed countries. Objectives: The aim of the present study was to determine the spatial factors related to the frequency of traffic crashes on pedestrians in districts of Tehran city. Methods: This was a cross-sectional study. All traffic crashes on pedestrians during 2013–2015 were included in this study. The data were extracted from different sources such as traffic police main office, Tehran municipal office, and Iran statistics center. Poisson and negative binomial regression models were used to analyze the role of environment and district on frequency of traffic crashes. Moreover, the likelihood-ratio test has been used for comparison between models, while assessment of goodness-of-fit has been reported using R2, Akaike information criteria, and Bayesian information criteria. Results: Of 12,090 crashes, 11,895 (98.4%) had led to injuries, while 195 (1.6%) had led to deaths. The frequency of crashes varied substantially in different districts of Tehran. The incidence of injuries did not show any particular pattern, whereas the pattern of incidence of deaths was lower in central districts in comparison to higher incidence in marginal districts (e.g., north, south, west, and east of Tehran). The results of the final model showed a statistically significant association among various variables such as demographics, web of roads, rate of traveling, and land use with the outcome as number of crashes in geographic units. Conclusions: Frequency distribution of traffic crashes leading to injury and/or death is completely different in various districts of Tehran. Demographic as well as spatial characteristics also play an important role in determining this distribution. Regional planning, appropriate traffic management, control measures on spatial risk factors, and educational programs could substantially improve the safety of pedestrians in Tehran.

Keywords: Injuries, pedestrians, spatial analysis, traffic


How to cite this article:
Moradi A, Soori H, Kavousi A, Eshghabadi F, Rahmani K, Nematollahi S, Zeini S. Spatial factors related to traffic crashes on pedestrians in all districts of Tehran. Arch Trauma Res 2018;7:133-9

How to cite this URL:
Moradi A, Soori H, Kavousi A, Eshghabadi F, Rahmani K, Nematollahi S, Zeini S. Spatial factors related to traffic crashes on pedestrians in all districts of Tehran. Arch Trauma Res [serial online] 2018 [cited 2024 Mar 28];7:133-9. Available from: https://www.archtrauma.com/text.asp?2018/7/4/133/259505


  Background Top


Traffic crashes result in 1.35 million annual deaths and cause injuries to tens of millions of people around the world.[1] Pedestrians are regarded as the most vulnerable users of roads. Substantial portion of fatal traffic crashes is attributable to pedestrians in the developed world. In 2013, 22% of all fatalities as a result of road traffic crashes in the European Union were associated with pedestrians.[2] In 2010, 4280 deaths and 70,000 injuries occurred among pedestrians in the United States due to traffic crashes.[3] Each year, there are 400 pedestrians killed by road traffic crashes in Canada, accounting for 13% of all traffic casualties.[4]

Proportion of deaths related to traffic crashes grows larger in dense cities. In Mumbai and Delhi, for example, pedestrians accounted for 78% and 53% of traffic fatalities, respectively.[5]

There has been an increase of inner and outer city traffic burden during recent years due to the increase of automobile manufacturing in Iran, whereas the passages have not been equivalently expanded. Therefore, a large proportion of traffic accidents can be attributable to spatial and environmental factors such as quality of roads and passages.[6]

According to traffic police main office, fatal traffic crashes on pedestrians account for 40% of total traffic deaths in Tehran city.[7]

Assessment of geographical frequency distribution and severity of pedestrian crashes have been considered important, leading to adaptation of geographical and statistical analytic methods worldwide.

Iran has experienced a growing traffic burden due to increased automobile manufactures, recently; however, the roads have not been developed accordingly. Due to the impact of spatial factors on traffic crashes and intensification of crashes by low capacity of roads, investigation of the effects of spatial factors on incidence of traffic crashes is indeed imperative. Results of such investigations could be used to control plans more appropriately and target the reduction of these effects.

Tehran, the capital city of Iran, with 612 kg2 of area, is located in the north of the country. The population of Tehran according to 2016 census was 8,693,706,[8] while the number of stationary population is estimated to be 8,089,766 now. Tehran, which is the center of business and industry in Iran, is divided into 22 districts, 134 counties including Tajrish and Rey, and 560 traffic zones.[9] Daily, there are 15–19 million interurban travels which hamper traffic control strategies.[10] Construction of highways, development of public transportation systems, and different traffic legislations are part of such strategies. Development of subway roads was another strategy which was responsible for transport of 129 million passengers on 2011.[11] Tehran is ranked as the 25th populated city worldwide and is estimated to become more populated in 2030.[12]

Objectives

The aim of the present study was to determine the spatial factors related to the frequency of traffic crashes on pedestrians in districts of the city of Tehran.


  Methods Top


This is an ecological study on casualties of traffic crashes on pedestrians during 2013–2015 in spatial units. Different data sources such as traffic police main office, Tehran municipal office, and Iran statistics center had been used. The dependent variable (outcome) was the rate of injuries or deaths in any district. The definitions of independent variables are provided in [Table 1].[9]
Table 1: Definition of independent variables

Click here to view


Statistical analysis was done using computer software such as SPSS Version 2 0, (SPSS Inc., Chicago, Ill., USA), STATA Version 12, (College station, Texas 77845, USA) and ArcGIS 10.2.2, (ESRI Redlands, USA). Descriptive statistics (e.g., mean, standard deviation [SD], and range) as well as maps were used to describe the distribution of crashes. Regression models including negative binomial (NB) regression and generalized Poisson regression were used for analysis of the effects of spatial and regional factors on frequency of traffic crashes on pedestrians. Unit of analysis was every district of Tehran metropolitan.

Given the statistically significant difference of overdispersion index with 0 in nonfatal crashes (P< 0.001), the use of NB regression model was justified, while nonsignificant difference in fatal events (P = 0.5) justified the use of generalized Poisson regression model.

We used maximum likelihood methods for estimation of regression coefficients and likelihood-ratio tests for comparisons of the models. We also assessed the goodness-of-fit models using Akaike information criteria and Bayesian information criteria.[13],[14]

First, if the P value of variables was < 0.2 in the univariate analysis, they were included in the model. Then, various models were compared to determine which one was the best in terms of goodness-of-fit.


  Results Top


Among 12,090 crashes being studied, 11,895 (98.4%) nonfatal and 195 (1.6%) fatal crashes were reported. The frequency distribution of nonfatal and fatal crashes was quite different in Tehran city. The absolute number of fatal crashes ranged between the lowest in central districts with one case per districts coded 11 and 12 and the highest with 27 cases in district coded 2 and 22 cases in district coded 4 [Figure 1].
Figure 1: Frequency of injuries of pedestrian-related crashes in districts of Tehran 2013–2015

Click here to view


On the other hand, the incidence of nonfatal crashes did not follow any particular pattern. However, this number was higher in central districts compared to marginal ones. For instance, the district coded 6 with 1092 cases and the district coded 5 (located in the north-west) with 1018 cases had the highest number of cases, followed by districts coded 2 and 12 with more than 750 cases, while districts coded 17 and 22 had the lowest number of cases [Figure 2].
Figure 2: Frequency of fatal pedestrian-related crashes in districts of Tehran 2013–2015

Click here to view


Furthermore, the incidence rate of injury and fatal crashes was quite different in Tehran city. The incidence rate of fatal crashes ranged between the lowest in central districts and the highest in southern districts. In contrast, the incidence rate of injury crashes ranged between the lowest in marginal areas of Tehran, especially in the southern and eastern districts, and the highest in central districts [Figure 3].
Figure 3: Incidence of injury and fatal pedestrian-related crashes in districts of Tehran 2013–2015

Click here to view


There were 3 different categories of variables, namely population and exposure, web of roads, and traveling and land use, which were included to investigate their relationship with traffic crashes on pedestrians in all 22 districts of Tehran. [Table 2] demonstrates the descriptive characteristics (mean, SD, minimum, and maximum) of the independent variables.
Table 2: Descriptive statistics of independent variables

Click here to view


According to final NB regression model, frequency of nonfatal crashes on pedestrians had a significant association with population density, percentage of illiteracy, percentage of employees at workplace, percentage of students at residential place, length of arterial roads to length of all roads, lack of parking space, educational land used to total area, and total land used to total area.

In addition, population density, percentage of illiteracy, percentage of employees at workplace, percentage of students at residential place, length of arterial roads to length of all roads, and educational land used to total area indicated a positive association, whereas lack of parking space and total land used to total area indicated a negative association with outcome variable [Table 3].
Table 3: Pedestrian injury crashes final negative binomial model for districts in Tehran

Click here to view


According to final generalized Poisson regression model, frequency of fatal crashes on pedestrians had a significant association with percentage of employees at workplace, length of highways to length of all roads, length of arterial roads to length of all roads, educational land use, industrial land use, and transportation as well as depository land use.

In addition, length of highways to length of all roads, length of arterial roads/length of all roads, educational land use, and industrial land use indicated a positive association, whereas percentage of employees at work and transportation as well as depository land use indicated a negative association with outcome variable [Table 4].
Table 4: Pedestrian fatal crashes final generalized Poisson model for districts in Tehran

Click here to view



  Discussion Top


According to [Figure 1], marginal districts including north-west, east, south, and south-east of Tehran were among the high-risk areas for fatal crashes on pedestrians.

[Figure 2] showed that districts coded 6 in the center and 5 in the north-west had the highest number of nonfatal crashes with more than 500 events annually. Being the center and ranked as the 4th dense district (10,240 habitants in square kilometers) in Tehran, district coded 6 has experienced much severe nonfatal traffic crashes that could be explained by the huge burden of transportation and commuters (952,984 commutes daily)[9] and smaller area compared to district coded 5.

Overall, there were different patterns of fatal and nonfatal traffic crashes in 22 districts of Tehran during 2013–2015. The present study revealed a significant negative association between population and employees' densities with fatal traffic crashes. In other words, fatal crashes occurred less commonly in districts with low population densities Sand lower percentage of employees at work. On the other hand, the frequency of nonfatal traffic crashes was significantly positively related to population characteristics such as population density, percentage of illiteracy, percentage of employees at work, and percentage of students at residence. It can be stated that nonfatal traffic crashes occurred more frequently in districts with high population density and higher percentage of employees at work.

Several studies including Cottrill and Thakuriah,[15] Sebert Kuhlmann et al. in Denver city, Colorado,[16] Siddiqui et al. in Florida,[17] Ukkusuri et al. in New York,[18] and McArthur et al. in Michigan[19] have found a significant relationship between population density and frequency of traffic crashes on pedestrians. Although the direction of their findings was the opposite of what we have found in fatal crashes, increment in number and severity of crashes in marginal districts can be explained by the fact that lower population density and percentage of employees at work decrease the traffic burden in roads and consequently lead to increase the pace of traffic. Other explanations related to cultural and safety equipment differences between Tehran compared to American cities.

The proportion of length of highways to total roads which is an important component of transportation had a positive significant relationship with the frequency of fatal crashes. In other words, the incidence of fatal crashes was higher in areas with higher highway densities. Speedy motor vehicles in such roads can possibly cause more severe and fatal crashes on pedestrians. In addition, in area level, frequency of nonfatal crashes had a significant positive relationship with transportation variables such as length of arterial roads and a negative relationship with the lack of parking and per capita ownership of private cars.

In other words, the incidence of nonfatal crashes was much higher in areas with more arterial roads. This finding is consistent with findings of several other studies, for example, Hashimoto in Florida,[20] Cloutier et al. in Canada,[21] and Green et al. in England.[22] Therefore, unsafe arterial roads can be considered as one of the reasons for increasing the incidence of nonfatal traffic crashes on pedestrians. Therefore, improving safety of arterial roads is necessary.

Distribution of fatal crashes had a significant relationship with land use variables such as educational, industrial, transportation, and depository land use in a way that the incidence of fatal crashes was higher in areas with high density of industrial and educational centers. High traffic rate of heavy automobiles through these areas could be one possible explanation for this phenomenon.

Distribution of nonfatal crashes was also related to land use variables such as total land use and educational utilization. The incidence of such crashes was higher in educational areas, which implied unsafe commutes for students.

It is worth mentioning that one of the effective reasons in traffic crashes on pedestrians is land use as shown by Yao and Loo in Hong Kong,[23] Ukkusuri et al. in New York,[18] Wier et al. in the United States.[24] Therefore, incorporation of safety measures in development designs for reconstruction of old city structures is important.

Our results showed that increased level of development and the number of parks within 22 districts of Tehran was related to increased incidence of nonfatal crashes. However, this relation is not significant, particularly, more developed areas could be concomitant with ignorance of pedestrians' safety during development process, and the increase in numbers of parks could uptake more passengers through their leisure times. Hence, improving safety of pedestrians around parks could possibly prevent such crashes.

The present study had some limitations. Fatal traffic crashes are defined as any death related to motor vehicles within 1 month after the crash. For the purpose of this study, we used traffic police data which did not cover fatal crashes within the defined period of 1 month. Because of lack of unregistered geographical coordinates at the time of a crash, police officers use spot maps and approximate geographic coordinates to register the crash. Accordingly, the real place of crash could be slightly different from what was registered. Few cases with unreliable police-registered spots compelled us to make a contact with the casualty to get more accurate information. Nonetheless, determination of the accurate place of crash could still be misleading. Poisson modeling for our interesting outcome, crashes injury, was another limitation of the study because this model is pertinent for rare data.

It is possible to change spatial objects over time. To resolve this problem, we used the latest information of traffic police, and estimation was undertaken using geographical and time layers.

One of the strengths of our study was the systematic use of Tehran traffic police data to determine the influence of spatial factors on the distribution of fatal and nonfatal traffic crashes on pedestrians.

There are only a few studies investigating spatial factors on the crashes on pedestrians in Iran using geographic information system and NB and generalized Poisson regression models. In this study, we used geographical information system to characterize high-risk areas of traffic crashes, while regression models were used to determine effective factors influencing frequency distribution of those crashes. This study showed that depending on the availability of reliable information, regression models are a suitable choice for determining influential factors and using the findings to target preventive measures. According to our findings, for pedestrians, marginal districts such as north, east, south, and south-west of Tehran (coded as 5, 4, 3, 15, 19, 20, and 21) are among the high-risk areas for fatal crashes, while central and north-west districts of Tehran (coded as 5 and 6) are among high-risk districts for nonfatal crashes. Frequency of fatal and nonfatal crashes on pedestrians is related to different spatial factors, especially population density, percentage of students, length of highways and arterial roads, and land use. More suitable designation of preventive programs for safety of citizens and decrease of traffic crashes are highly recommended. It is imperative to assess these factors more carefully and adapt applicable approaches to modify them if possible. It should be noted that some effective interventions such as designing safer infrastructure, improving the safety features of vehicles, speed restriction in crowded places, improving postcrash care for victims of road crashes, setting and enforcing laws relating to key risks, and raising public awareness can be used to reduce traffic crashes.


  Conclusion Top


According to the results, there are various high risk districts for pedestrians fatal crashes in Tehran. North, east, south and south west are among the high risk areas for fatal crashes, while central and north west districts are among high risk districts for nonfatal crashes. Demographic and spatial characteristics play an important role in different distribution of traffic crashes leading to injury and/or death. It concluded that regional planning, appropriate traffic management, control measures on spatial risk factors, and educational programs could substantially improve the safety of pedestrians in Tehran.

Acknowledgment

The authors are grateful for kind assistance of School of Public Health authorities, faculty members of Epidemiology Department, and traffic police of Rahvar Tehran authorities. We would like to sincerely thank all who helped us in this study, especially head of crashes office of Tehran traffic police, Colonel Salah Al-Din Zaini.

Financial support and sponsorship

This study conducted with the cooperation of the Department of Epidemiology of Shahid Beheshti University of Medical Sciences and traffic police of Rahvar Tehran. This project was funded by Shahid Beheshti University of Medical Sciences.

Conflicts of interest

There are no conflicts of interest.

 
  References Top

1.
World Health Organization. Global Status Report on Road Safety 2018. Licence: CC BYNC-SA 3.0 IGO. Geneva: World Health Organization; 2018.  Back to cited text no. 1
    
2.
European Union, Road Safety in the European Union. Trends, Statistics and Main Challenges. European Union, Road safety in the European Union; 2015. Available from: http://ec.europa.eu/roadsafety. [Last accessed on 2015 Feb 01].  Back to cited text no. 2
    
3.
Slaughter DR, Williams N, Wall SP, Glass NE, Simon R, Todd SR, et al. Acommunity traffic safety analysis of pedestrian and bicyclist injuries based on the catchment area of a trauma center. J Trauma Acute Care Surg 2014;76:1103-10.  Back to cited text no. 3
    
4.
Miranda-Moreno LF, Morency P, El-Geneidy AM. The link between built environment, pedestrian activity and pedestrian-vehicle collision occurrence at signalized intersections. Accid Anal Prev 2011;43:1624-34.  Back to cited text no. 4
    
5.
Mohan D, Tsimhoni O, Sivak M, Flannagan MJ. Road Safety in India: Challenges and Opportunities. The University of Michigan, Transportation Research Institute; 2009.  Back to cited text no. 5
    
6.
Moradi A, Soori H, Kavousi A, Eshghabadi F, Nematollahi S, Zeini S. Effective environmental factors on geographical distribution of traffic accidents on pedestrians, downtown Tehran city. Int J Crit Illn Inj Sci 2017;7:101-6.  Back to cited text no. 6
[PUBMED]  [Full text]  
7.
Accident Bureau. Bulletin of Statistics and Analysis of Accidents. Tehran: Traffic Police of Rahvar; 2015.  Back to cited text no. 7
    
8.
Iran's Population and Housing Census-2016. Iran Statistics Center; 2018.  Back to cited text no. 8
    
9.
Center for Studies and Planning in Tehran. Mobility, Transportation and Communication Networks of Tehran Comprehensive Plan. Ministry of Housing and Urban Development and Tehran Municipality; 2006.  Back to cited text no. 9
    
10.
Tehran Municipality. Tehran; 2015. Available from: http://www.tehran.ir. [Last accessed on 2015 Feb 01].  Back to cited text no. 10
    
11.
The Number of Trips Made. Teran: Railway Company of Tehran and Suburbs; 2011.  Back to cited text no. 11
    
12.
United Nations, Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2012 Revision. New York, USA: United Nations, Department of Economic and Social Affairs, Population Division; 2013.  Back to cited text no. 12
    
13.
Ayati E, Abbasi E. Application of zero-inflated regression models in modeling accidents on urban highways. Omran Modares J 2011;11:1-15.  Back to cited text no. 13
    
14.
Hilbe JM. Negative Binomial Regression. London, England: Cambridge University Press; 2011.  Back to cited text no. 14
    
15.
Cottrill CD, Thakuriah PV. Evaluating pedestrian crashes in areas with high low-income or minority populations. Accid Anal Prev 2010;42:1718-28.  Back to cited text no. 15
    
16.
Sebert Kuhlmann AK, Brett J, Thomas D, Sain SR. Environmental characteristics associated with pedestrian-motor vehicle collisions in Denver, Colorado. Am J Public Health 2009;99:1632-7.  Back to cited text no. 16
    
17.
Siddiqui C, Abdel-Aty M, Choi K. Macroscopic spatial analysis of pedestrian and bicycle crashes. Accid Anal Prev 2012;45:382-91.  Back to cited text no. 17
    
18.
Ukkusuri S, Miranda-Moreno LF, Ramadurai G, Isa-Tavarez J. The role of built environment on pedestrian crash frequency. Saf Sci 2012;50:1141-51.  Back to cited text no. 18
    
19.
McArthur A, Savolainen P, Gates T. Spatial analysis of child pedestrian and bicycle crashes: Development of safety performance function for areas adjacent to schools. Transportation Research Record. J Transp Res Board 2014;2465:57-63.  Back to cited text no. 19
    
20.
Hashimoto T. Spatial Analysis of Pedestrian Accidents. USA: University of South Florida; 2005.  Back to cited text no. 20
    
21.
Cloutier MS, Apparicio P, Thouez JP. GIS-based spatial analysis of child pedestrian accidents near primary schools in Montréal, Canada. Appl GIS 2007;3:1-18.  Back to cited text no. 21
    
22.
Green J, Muir H, Maher M. Child pedestrian casualties and deprivation. Accid Anal Prev 2011;43:714-23.  Back to cited text no. 22
    
23.
Yao S, Loo BP. Identification of hazardous road locations for pedestrians. Procedia Eng 2012;45:815-23.  Back to cited text no. 23
    
24.
Wier M, Weintraub J, Humphreys EH, Seto E, Bhatia R. An area-level model of vehicle-pedestrian injury collisions with implications for land use and transportation planning. Accid Anal Prev 2009;41:137-45.  Back to cited text no. 24
    


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4]


This article has been cited by
1 Clustering and pedestrian crashes prediction modelling: Amman case
Lina Shbeeb
International Journal of Injury Control and Safety Promotion. 2023; : 1
[Pubmed] | [DOI]



 

Top
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Background
Methods
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed5089    
    Printed412    
    Emailed0    
    PDF Downloaded483    
    Comments [Add]    
    Cited by others 1    

Recommend this journal