EPIDEMIOLOGY OF ROAD TRAFFIC CRASHES REPORTED IN THE KURUNEGALA POLICE DIVISION IN SRI LANKA

Background: Road Traffic Crashes (RTCs) kill 1.24 million people in the world annually. It has become a major but neglected public health problem in the world including Sri Lanka. Objective: To describe characteristics of RTCs and socio-demographic characteristics of involved road users reported at the Kurunegala Police Division (KPD), Sri Lanka during April to December in 2013. Materials and Methods: A descriptive cross-sectional study was conducted and the study participants were involved road users of consecutively reported RTCs to ten police stations in the KPD. Data were collected using an interviewer administered structured questionnaire and a data record sheet. Results: 851 RTCs were reported during the study period with 1481 vehicles and 1887 road users involved. 7.8% of the RTCs were fatal, 70% resulted in non-fatal injuries and 22.2% caused damage only. The average rate of RTCs was 03 per day and majority (37.3%) was reported from urban areas between 12.00hr-17.59hrs. Fatal RTCs were higher between 0.00hr-05.59hrs in rural areas. The leading type of vehicles involved were motorcycles and the most vulnerable road users were males (84%) between 3140 years. The majority of motorcyclists and pillion riders (82.8%) were wearing a helmet at the time of the crash. The majority of pedestrians (85.2%) were injured while crossing the road, out of which 34% were on a pedestrian crossing. Conclusion: Young males in productive age were the most affected by RTCs.


Introduction
More than 1.24 million people die from Road Traffic Crashes (RTCs) annually and about 90% of road traffic fatalities occur in low-income and middle-income countries, which comprise two-thirds of the global population 1 . RTCs are one of the top five causes of morbidity and mortality in the South Asian region 2 . RTCs have become an important public health problem that needs urgent attention.
Sri Lanka, a middle income country, is facing the burden of RTCs, and associated injuries and death, due to an exponential growth in motorization on a static road system. The traffic Police statistics of Sri Lanka shows an increase of RTCs reported during the recent past. During 2009-2015 period, the total number of RTCs increased by 15% while deaths increased by 17% 3 .
The North Western Province (NWP) of Sri Lanka consists of six police divisions including the Kurunegala Police Division (KPD). This study was conducted in the KPD which is the most populated police division in the NWP, with an estimated mid-year population of 300,872 and a population density of 466 persons per km 2 in 2011 4 . In addition to the residential population, around 75,000 people visit the Kurunegala city, the capital of NWP, daily for education, administrative and business purposes 4 . Kurunegala is a major transitional centre in Sri Lanka, connecting Colombo, the capital with the other parts of the country. This has made Kurunegala a city with a high vehicle density, congestion and its flat, broad and straight road network conducive to high speed driving.
A comprehensive literature search on RTCs in the KPD revealed a scarcity of published articles, which made us to conduct this study to describe the epidemiology of RTCs reported in the KPD, Sri Lanka.

Materials and Methods
This descriptive cross-sectional study was conducted at the ten police stations and nine government hospitals situated in the KPD. A RTC was "an event that produced injury and/or property damage, involving a vehicle in transit, and occurring on a road" 5 and an Involved Road User (IRU) of a RTC was "a person who was driving or was in a vehicle involved in the RTC or a pedestrian hit by a vehicle" 1 .
All the RTCs reported to the ten police stations in KPD from April to December in 2013 and all IRUs were included in the study. IRUs who, did not go to a police station or were not admitted to a government hospital situated in KPD were not included. With an expected proportion of injury producing RTCs of 55% 6 and a 3.5% desired precision, a sample size of 775 RTCs were obtained and with an expected non-response rate of 10%, 851 RTCs were studied 7 . A pre-tested, interviewer administered structured questionnaire and a data record sheet were used for data collection and young male trained data collectors with similar educational background collected data.
Any RTC needs to be reported to a police station within a stipulated period from the time of the crash, as specified in the Motor Traffic Act of Sri Lanka 8 . All injured IRUs admit to a government hospital, especially if the injured expect to initiate legal action and file for compensation. As health care at government hospitals in Sri Lanka are free of charge there is no inhibition to request admission to a hospital. In this study, the data collector was immediately informed of the RTC by the police officer attending the crash, and then the data collector visited the relevant police station or the hospital and interviewed the IRU. If the IRU was severely injured and the condition did not warrant an interview, his/her socio-demographic data was obtained from the relatives. When the relevant information could not be obtained at the first visit, it was done at a subsequent visit/s. The information of the injured IRU who succumbed to their injuries was collected from the closest relative at the postmortem.
Ethical clearance for the study was obtained from the Institutional Ethical Review Committee, Faculty of Medicine, University of Peradeniya, Sri Lanka and permission to conduct the study was obtained from relevant authorities. Informed written consent was obtained from the study participants prior to the data collection. Data was cleaned, entered to an Excel data sheet and imported to SPSS and analyzed using SPSS version 21. A univariate analysis was conducted initially and for selected variables a bivariate analysis was conducted subsequently, chi-squared test and student t test were used to identify statistical significance. A p value of <0.05 was used to identify statistical significance.

Results
The study included 851 RTCs involving 1481 vehicles and 1887 road users during nine months study period.
The majority (77.8%) of the RTCs caused an injury to an IRU. Three RTCs per/day were reported. Majority of fatal crashes occurred on public holidays (14%). A statistically significant association was observed between the time of the day and the severity of the RTC (x2=23.293, df=6, p<0.05) ( Table 1).
Occurrence of all types of RTCs (84.1%,n=716) and the fatal RTC percentage (8%) were higher on major roads than on minor roads. A statistically significant association was observed between the severity of RTC and the type of the junction where it occurred (x 2 =145.5637, df =4, p < 0.05) ( Table 2).
The majority of drivers/motorcyclists had a valid driving license (81.4%,n=1150) and the majority of motorcyclists and pillion riders (82.8%) were wearing a helmet. The majority of pedestrians (144, n=85.2%) were injured while crossing the road, out of which 34% were on a pedestrian crossing (Table 5).

Discussion:
Type of RTCs: A total of 1481 vehicles and 1887 road users were involved in 851 RTCs in the KPD during the nine months in 2013, with 7.8% fatal, 70.0% non fatal injury-producing crashes. These figures were higher than the corresponding national figures (6.1% for fatal and 49.1% for non fatal injury producing) in 2011 indicating an increased mortality and morbidity 6 . An average of 95 RTCs per/month was reported. This was similar to the average in the study area during 2008 to 2011 (100 per/month) 4 . The recorded rate of RTCs was similar on a week day, week end or a public holiday which was 3 RTCs per day in KPD. More

Socio demographic characteristics of the IRUs:
Nearly 84% of IRUs were males and 33.4% of them were in the 31-40 year age group. The majority of drivers (88.4%), and motorcyclists (97.3%) were males. The male predominance among drivers confirms the traditional and cultural aspect present in the country and observed in previous studies in Sri Lanka, showing that more than 80% were men 13 . In the present study, drivers and motorcyclists consisted the majority (75%), with similar statistics reported in 2012 in Sri Lanka 6 and in a study conducted in Batticoloa 15 . Involvement of pedestrians (9.3%) and pedal cyclist (3.5%) were low in the present study. A study conducted in 2005 highlighted that pedestrians/pedal cyclists accounted for more than 50% of all RTCs and motorcyclists accounted for an additional 13% 16 . The changes in economic growth in Sri Lanka may have contributed for the different findings, with more people using vehicles for transportation at present. Importantly, 17.0% of pedestrians were more than 60 years of age, highlighting the dependent nature of the road user. The majority of pedestrians (85.2%) were injured while crossing the road, out of which 34% were on a pedestrian crossing. These findings highlight the need for strict road safety measures in the study area. Driving needs skills and requires training and experience which is a crucial factor for road safety 1 . Having a valid license is considered as a compulsory requirement to drive a vehicle. The study revealed that the majority of drivers/motorcyclists had a valid driving license (81.4 %) at the time of the crash as seen in previous studies, conducted in Kandy (93.6%) and Batticoloa (83%) 13,15 . However, it was as low as 15% of drivers in an Indian study 9 . The WHO, stated that rapid increase of motorcycle use in low and middle-income countries is being accompanied by an increase in the incidence of head injuries. The non-use of protective helmets is a well established risk factor for head injuries 1 . The present study showed that the majority of motorcyclist and pillion riders were wearing helmets (82.8%) and the usage of helmets was significantly higher among motorcyclists than pillion riders. The helmet usage in the KPD among motorcyclists (87.3%) and pillion riders (33.3%) was lower than the corresponding national figures which were 99% and 87% respectively; in 2010 17 . A study conducted in Batticaloa reported that only 25% of riders/pillion riders were wearing a helmet 15 and zero use of helmets by the users of motorized two-wheelers in an Indian study 9 .

Road characteristics:
The percentage of RTCs recorded on major roads (84.1%) was higher than on minor roads which coincide with the road structure categorization of KPD, where the length of major roads is higher than the length of minor roads 17 . A majority of RTCs (80.7%) occurred on locations where there was no junction within 10 meters. Although the recorded RTCs was low (2.0%) on locations near roundabouts, entrance byroads and rail road crossings, the recorded fatal RTCs (29.4%) were higher on those locations followed by "junctions" (11.6%). A statistically significant association was observed between the severity of RTC and the type of the junction where it occurred. These findings reveal the behavior of the drivers where they tend to drive faster on straight roads. Disobeying the road rules may be the cause for higher fatal RTCs on locations near roundabouts, entrance byroads, rail road crossings and junctions.

Vehicular Characteristics:
Motorcycle (33.4%) was the main vehicle responsible for any type of RTC while 12.4% was by the three wheelers. The country profile of traffic statistics also showed that motorcycles and threewheelers were responsible for 38.7% of all RTCs in 2011 6 . RTCs caused by motorcycle was the commonest (71%) in Batticaloa 15 and was 22.8% in an Indian study 9 . The use of vehicle is frequently related to the socio-economic status of the road user 18 . Nearly 50% of motorcycles and 25% of three-wheelers were newly registered in the year 2012 19 in Sri Lanka and they are the cheap and affordable transportation modes although they are comparatively vulnerable vehicles in the aspect of safety to the user. Overcrowded, unsafe modes of public transport also contribute to RTCs in Sri Lanka. The public transport buses are either owned by the government or by private owners. The private-owned buses (5.3%) were more involved in RTCs than government buses (1.8%) indicating that travelling in a privately-owned bus has a higher risk of RTCs than travelling in a government bus as also shown in a previous study conducted in Kandy 20 .

Limitations of the study:
As specified in the Motor Traffic Act of Sri Lanka 8 , all RTCs should be reported to the nearest police station. Although the law mandates this, all RTCs which occurred in the KPD, may not have been reported to a police station. Also, all injured IRUs may not have visited a government hospital in the KPD. These persons were not included in the study.

Conclusions and Recommendations
The study concluded that young productive males were the most vulnerable IRUs and the motorcycle was the highest involved vehicle in RTCs.
Interventions in RTCs and road traffic injury prevention are broad based and include rules and regulations, legal action and community awareness. In primary preventive measures of RTCs, the focus should be on implementation of clientoriented community and school-based awareness programmes on road signs, pedestrian's and other road users' rights, their behaviors and the use of protective measures, such as helmets and seat belts. Severity of injuries can be reduced by improving quality of pre-hospital care and improving the knowledge of first-aid. Strict implementation of the helmet law for motorcycle riders should be done by the Department of Police. Identification of "hot spots" and CCTV cameras will be useful to reduce RTCs and the severity of injuries. Attention should be paid to the implementation of road safety measures for road users in urban development and roadway planning.
Young productive males required for development in the country are unnecessarily being killed and injured by RTCs. The RTCs, and associated mortality and morbidity may be contributing to the propagation of poverty in the country by injuring the principle wage earner of the family and RTCs have become a social problem too. Therefore, prevention and control of RTCs and associated injuries and deaths are an urgent need for Sri Lanka.

Contribution by authors:
The planning of the research study, collection of data, data analysis and writing of the manuscript done by P.G. Amarasinghe. Supervising and evaluation of the research project was done by Prof. S.D. Dharmaratne.
No conflict of interest of this research study.