Bulletin of the World Health Organization

Trends in traffic collisions and injuries in Kyrgyzstan, 2003–2007

Viola Artikova a, Michael E Thompson b, Elena Platonova b, Gerald F Pyle b & Samat Toimatov c

a. The Resource and Policy Exchange, Tashkent, Uzbekistan.
b. Department of Public Health Sciences, University of North Carolina at Charlotte, 9201 University City Boulevard, Charlotte, NC, 28223-001, United States of America.
c. Division of Health Care Services, Ministry of Health of Kyrgyzstan, Bishkek, Kyrgyzstan.

Correspondence to Michael E Thompson (e-mail: methomp1@uncc.edu).

(Submitted: 10 November 2010 – Revised version received: 18 March 2011 – Accepted: 21 March 2011 – Published online: 30 March 2011.)

Bulletin of the World Health Organization 2011;89:345-351. doi: 10.2471/BLT.10.084434


Motor vehicle collisions are a neglected public health problem. Approximately 45% of all hospitalized trauma patients in the world have been injured in a motor vehicle collision1 and an estimated 1.3 million people die in collisions annually.24 Traffic collisions are among the 10 leading causes of death worldwide and are projected to become the third leading cause of disability-adjusted life years (DALYs) lost by 2020.5 Adolescents and young adults (people aged 10–24 years) are at greater risk of being involved in motor vehicle collisions than older people4,68 and men are at greater risk than women.1,6,9 Motor vehicle collisions are also the leading cause of injury among children.9 In low-income countries, children are most often injured in collisions involving minibuses, buses, trucks and other commercial vehicles.9,10 The direct cost of motor vehicle collisions is roughly equivalent to 1% of the gross national product (GNP) in low-income countries and to 1.5% and 2% in middle- and high-income countries, respectively.5,11,12 Although not all factors contributing to motor vehicle collisions are modifiable, many of them are. These include environmental factors such as road planning and traffic infrastructure, speed limits and the state of repair of driven vehicles, as well as human factors such as driving style and skills, state of alertness on the road, and the use of alcohol or drugs while driving.3,12,13

In many low-income countries, lack of money and human resources has impeded proper action to reduce motor vehicle collisions and injuries and mortality from traffic injuries has increased rapidly.3,1416 However, the declines observed in most high-income countries in traffic collisions and in road traffic injuries and deaths suggest that these problems and the resulting morbidity and mortality can be reduced in low- and middle-income countries as well.16

Nantulya & Reich17 identified four reasons that explain why developing countries face a higher burden of motor vehicle collisions and injuries than developed countries: a proportionately higher increase in the number of motor vehicles, poor enforcement of traffic safety regulations, inadequate public health infrastructure and human resource capacity, and poor health-care services. Developing countries also have inadequate systems for collecting data on motor vehicle collisions and injuries and more fragmented response mechanisms. Consequently, both collisions and casualties are underreported.12 Several studies conducted in low- and lower-middle-income countries have shown an association between the number of vehicles per capita and the number of motor vehicle collisions and injuries.3,18,19 Furthermore, in low- and lower-middle-income countries, the reliability of the data aggregated by law enforcement and health agencies depends to some extent on the nature and severity of the injuries and their outcomes.7 For example, police records might more accurately record fatal traffic injuries, whereas hospital data might be more reliable in tracking non-fatal injuries. It is possible that neither captures collisions involving minor injuries.


Located in central Asia, Kyrgyzstan is a former Soviet republic. Approximately 90% of its territory is mountainous.20,21 Its population of over 5.3 million22 is predominantly rural (65%) and over 30% is composed of children and adolescents. With a human development index of 0.604 (122nd among 179 countries) 2006, Kyrgyzstan is classified as a low-income developing country.23,24

The standardized rate of death from motor vehicle collisions in Kyrgyzstan remains higher than for the region of central Asia as a whole. It increased by more than 60% between 2000 and 2005 and reached 17.43 per 100 000 people in 2006.25 Kyrgyzstan’s traffic fatality rates are 30 times higher than those of western Europe and three times higher than eastern Europe and Asia.12 The increasing number of traffic collision injuries and the relatively high rate of fatalities in Kyrgyzstan have raised concern.

In 2002, unintentional injuries were the second leading cause of disability in Kyrgyzstan among males and the sixth among females.26 In a 2008 interview, the head of the Department of Traffic Safety (DTS) stated that about 88% of motor vehicle collisions in the country occurred because of traffic rule violations.27 Besides lack of knowledge of traffic rules, alcohol intoxication among pedestrians and drivers’ impunity from prosecution were cited as contributing factors.28

The Kyrgyz National Statistical Committee29 reported over 208 000 privately-owned cars in 2007, equivalent to 40 per 1000 persons. Another source reported that in 2008 the number of registered cars had reached almost 425 000 after a 33% increase during the first 10 months of the year.30 These more recent data, if correct, bring up to 80 the number of vehicles per 1000 persons. In the light of rising collision and injury rates and the increasing numbers of automobiles in Kyrgyzstan, public health advocates need to raise public awareness of this important issue and inform policy-makers.

This study assessed the epidemiologic data on motor vehicle collisions and injuries in Kyrgyzstan to facilitate the development of evidence-based policy for the country.


This descriptive epidemiological study of secondary data analysed motor vehicle collision and injury trends in Kyrgyzstan between 2003 and 2007 by reconciling law enforcement and medical system data. A literature review has shown that relying solely on data from law enforcement agencies may be misleading because minor collisions often go unreported. Furthermore, in remote areas lacking ambulance services, victims often seek private transport to a medical facility before a traffic officer arrives on the scene.31 We supplemented the official statistics with information and insights obtained through personal interviews with health and law enforcement professionals.

Two distinct national reporting systems were our sources of data for the national, regional and (where available) facility level. Law enforcement agencies report vehicle registrations, motor vehicle collisions, deaths, injuries and related demographic information by region. This information is aggregated and reported by the Department of Traffic Safety, which is housed within the Ministry of the Interior and oversees the traffic police force. The Ambulance Services Department, a unit of the Ministry of Health, reports data on the number and types of casualties (the dead and the injured) and related demographic information and on the disposition of the casualties. Hospitals and outpatients clinics, also units within the Ministry of Health, report detailed statistics on trauma victims treated at those facilities. These health services data were aggregated and reported by the Ministry of Health’s Republican Medical Information Centre (RMIC). Our requests for information from private ambulance services went unanswered.

We organized and tabulated the data into a single Excel spreadsheet for descriptive and bivariate correlational analyses. In several cases, official agency data were internally inconsistent. When this occurred, we used the total supported by the more detailed categorical breakdown. Similarly, the totals for the same measure varied slightly by reporting agency, and in these cases we used the number reported by the agency with primary reporting responsibility for that statistic.


Law enforcement data

According to the Secretariat of Traffic Safety (source materials available from corresponding author), under the Commission on Ensuring Traffic Safety of the Government of Kyrgyztan, the number of vehicles in Kyrgyzstan increased by 25% between 2003 and 2007 and had reached nearly 320 000 by 2008. The number of collisions, deaths and injuries outpaced the growth in population and the number of vehicles on the road. In 2007, 4692 motor vehicle collisions were registered in Kyrgyzstan. Among them were 995 fatal collisions (21% of all collisions) that resulted in more than 1250 deaths and 6223 injured people (Table 1). These data reflect a 39% increase in the number of reported traffic collisions since 2003, a 40% increase in the number of road traffic deaths and a greater than 50% increase in the number of road traffic injuries.

According to the Department of Traffic Safety, pedestrians comprised the largest portion of those killed in motor vehicle collisions (approximately 45%), followed by passengers (approximately 30%), drivers (approximately 20%) and cyclists (2%). Passengers constituted the largest proportion of those injured (approximately 45%), followed by pedestrians (approximately 35%), drivers (approximately 20%) and cyclists (approximately 1%). Adults aged 25–40 years had the highest rates of injuries and deaths. The number of traffic collisions, injuries and deaths among children (16 years old or younger) increased by roughly 35% between 2003 and 2007. Drivers were at fault in over 70% of all motor vehicle collisions and pedestrians were responsible in 20%. Motor vehicle collisions caused by drivers increased by 42% between 2003 and 2007, whereas collisions caused by pedestrians increased by 21%. The data from the Department of Traffic Safety did not demonstrate a statistical association between the number of vehicles per capita and the number or severity of traffic collisions and injuries (data not shown).

Ambulance service data

State health-care institutions typically provide ambulance services in Kyrgyzstan. Four privately-owned ambulance services operate in Bishkek and Osh, the country’s two largest cities.31 As no reporting system exists for private sector ambulance services, only one of the private services reports data to the RMIC for inclusion in the official statistics reported here.31The number of calls for service due to traffic collisions increased by approximately 10% between 2004 and 2007, whereas the number of casualties served increased by over 65% and the number of deaths recorded, by nearly 60% (Table 2). The total number of calls received annually by the ambulance service remained relatively uniform during this period and averaged approximately 480 000 calls. The number of collision-related calls and of casualties served remained relatively uniform from 2004–2006, with most of the observed increases occurring from 2006 to 2007. However, collision-related deaths rose sharply from 2005 to 2006 and from 2006 to 2007. When we compared the number of motor vehicle collisions reported by the Department of Traffic Safety (Table 1) with the number of calls for ambulance services (Table 2), a widening gap between the two systems was noted.

Medical system data

The RMIC reported nearly 12 000 new collision-related injuries in 2007, an 18% increase over 2003 (source materials for data in this and the following paragraph available from corresponding author). Depending on the severity of their injuries, people involved in collisions are either sent to hospital or to an outpatient health-care institution known as a centre of family medicine. In 2007, nearly 62% of collision trauma patients were treated in hospitals and 38% were treated in centres of family medicine. Between 2003 and 2007, traffic-related injuries ranked fourth among causes of trauma in children 14 years of age or younger and third among those aged 15 and older. Traffic collisions were the leading cause of death from injuries among adolescents and adults and the second leading cause among children.

Between 2003 and 2007, the rate of mortality attributable to motor vehicle collisions increased by 39% and the overall case fatality rate increased by approximately 40% (from 5% to 7%). Case fatality rates at hospitals and centres of family medicine were similar but changes varied by age group. Among children the number of injuries decreased by 5% but deaths increased by 36%. The proportion of deaths attributed to motor vehicle collisions reported by the centres of family medicine increased by 36%, hospital-reported injuries by 35% and deaths by 17%. Among children 15 years of age and younger, the number of collision-related injuries and deaths reported by the centres of family medicine increased by 59% and 96%, respectively, whereas hospital data showed 2% and 26% increases, respectively. These medical system data suggest that traumatic injuries from motor vehicle collisions have increased by nearly one third in both relative and absolute terms.


Official data from law enforcement, ambulance and medical system sources point to a significant increase in the number of motor vehicle collisions and resulting casualties in Kyrgyzstan between 2003 and 2007. All increases outpaced the growth in population (approximately 4%) for the same period29 but were in closer proportion with the reported increase in vehicle registrations. Ambulance calls did not increase in proportion with the reported increases in vehicle collisions, perhaps because of reliance on private ambulance services. Still, the number of casualties among patients transported by ambulance increased considerably, as did the number of collision-related deaths overall.

In our study, the highest casualty rates were observed among pedestrians and passengers. This finding is consistent with those of a study conducted by Razzak et al.,9 who analysed injuries among children under 15 years of age in Pakistan. In their study, young adults (aged 25–40 years) had the highest rates of collision-related injuries and deaths; those killed were more likely to be pedestrians and those injured were more likely to be passengers. Drivers ranked third among the casualties and the injured.

The relatively larger increase in collision-related deaths and injuries in comparison with the increase in the number of collisions could be explained in several ways. More people may have been injured per collision; collision-related injuries may have been more severe; collisions not involving injuries may have been underreported; the effectiveness of the reporting system may have changed or the quality of the trauma care provided by ambulance services may have decreased. The role of seatbelt use and of the enforcement of this and other safety measures cannot be assessed based on the data.

The increase in the number of calls for ambulance services was proportionately lower than the increase in traffic collisions and injuries. This finding suggests that ambulance services were not accessible or were intentionally bypassed. The data do not provide insight into the underlying reasons for this discrepancy, but experience suggests that enhancing first-responder training could be beneficial.32 Although the analyses revealed no meaningful difference in fatality rates across regions or between centres of family medicine and hospitals, the data do not address the rate of preventable deaths, the degree to which patients were appropriately triaged (or self-triaged) to outpatient or inpatient facilities or the effects of access to and use of ambulance services. Correlation analyses did not suggest any relationship between rates of injury and death and population growth or the number of vehicles, either nationally or regionally. These findings are consistent with those of Garg & Hyder18 in India.

Our results should be interpreted in the light of several limitations. Official law enforcement data enumerated only reported incidents, which may not have included minor traffic collisions and injuries.10 Systematic under- or over-reporting of certain types of cases, casualties, risk factors or outcomes may have biased the results. Inconsistencies in data were identified within and across agencies. Clarifying inaccurate, missing or ambiguous data was not possible.

In addition, in Kyrgyzstan the system for documenting motor vehicle collisions and traffic injuries lacks consistency and data may be missing or ambiguous. The agencies involved in monitoring and reporting motor vehicle collisions and injuries use different reporting formats. Such factors make it difficult to compare operational data across institutions and agencies. In addition, anecdotal evidence shows that informal payments are commonly given to police officers and/or the injured party at the collision scene in exchange for not reporting minor offences.29,33 Such a practice, if widespread, can lead to an underestimation of less serious collisions.

Despite these limitations, our study is the first documented attempt to comprehensively analyse trends in motor vehicle collisions and injury patterns in Kyrgyzstan. By combining data from several agencies, it provides a more comprehensive assessment of the burden of traffic injuries than has so far been presented in the literature. More complete and accurate traffic injury data can better serve to raise public awareness and inform policy-makers of this increasingly important public health issue.

Conclusion and recommendations

Several general recommendations emerge from our findings and from the literature in connection with surveillance and research capacity and systems operation. First, data quality and consistency must be improved and data sharing across law enforcement, ambulance and medical service systems must be implemented and properly coordinated, as presently no single agency in Kyrgyzstan captures all traffic injuries and fatalities. Furthermore, differences in how injuries are defined, how surveillance is conducted and how motorists behave after a collision generate discrepancies in outcome estimates.

Second, more socioecological research is needed to fully understand the complex dynamics of motor vehicle collisions in Kyrgyzstan. The more expansive socioecological model, which emphasizes behavioural interventions and highlights the host as an advocate for cultural and policy change, is now being favoured over the Haddon matrix, often used in injury control.3436 Differences in rates of collision and injury among regions should be examined in the light of differences in environmental factors such as topography and road infrastructure, the degree of enforcement of traffic regulations, and socioeconomic status and sociocultural patterns, including driver behaviours and attitudes. A study of perceived driving skills and other individual-level factors may provide insight into how risky driving behaviours relate to driver education and licensure practices.

Good ambulance services are critical for reducing preventable deaths and improving the triage of trauma victims,15 and access to emergency medical services helps ensure appropriate trauma care for all. Law enforcement officers, who are the first to respond to traffic collisions, should be provided with comprehensive first aid training and emergency trauma kits. The Kyrgyz Red Crescent Society’s recent efforts to train police in basic first aid should be strengthened and expanded.37

In 2009 the Kyrgyz government began aggressively enforcing the use of seat belts by front-seat passengers,38 but the seat belt law does not apply to children under 12 or to rear seat passengers. Safety laws targeting children need strengthening, as children are at increased risk of injury or death in collisions.

Finally, poor road design and signage, compounded by an increase in traffic volume, contribute increasingly to vehicle collisions.3,12 The transportation network must be periodically evaluated and adapted in the light of changes in traffic mix and volume.

In conclusion, this study adds to the knowledge base by describing the pattern of motor vehicle collisions and injuries in Kyrgyzstan and documenting the growing need to address traffic safety in the country. It also highlights the need for a comprehensive, systematic and transparent traffic surveillance and monitoring system. Further research drawing on a socioecological framework is needed to provide a scientific base for policy discussions on how best to improve road safety.


At the time of the study, Viola Artikova was a graduate student in the Department of Public Health Sciences at the University of North Carolina in Charlotte.

Competing interests:

None declared.