Epidemiology of road traffic injuries and fatalities among children in Ethiopia: a systematic review and meta-analysis Authors: Mohammed Seid Ali, Angelo D’Elia, Jennie Oxley, Sjaan Koppel Published: 2026 ABSTRACT AbstractBackgroundRoad traffic injuries are a global public-health challenge, causing substantial morbidity and mortality, especially among children in low- and middle-income countries such as Ethiopia. Despite this impact, research on child road traffic injuries is often under-represented and often masked within the general population. Effective injury prevention requires context-specific evidence to support implementation and scale-up interventions. This study aimed to address these gaps by investigating epidemiological evidence and injury patterns among paediatric populations.MethodsA systematic literature search was conducted across major databases, including Medline, Embase, Scopus, PubMed, CINAHL Complete, and Ovid Transport to identify published studies reporting on child traffic injuries. Data were extracted using a standard Microsoft Excel format, and the quality of the included studies was assessed using the JBI checklists. Statistical analysis was conducted using Stata version 18. Random-effects models were used to estimate the pooled prevalence of child traffic injuries and fatalities.ResultsA total of 36 articles were included in this review. The random-effects models indicated that the pooled prevalence of child traffic injuries and fatalities was estimated at 23% (95% CI: 21%-25%) and 17% (95% CI: 11%-24%), respectively. Regarding the pattern of injuries, 81.5% of all reported injuries involved pedestrians; 40.6% of injury cases involved individuals aged between 13 and 18 years; 37% of injuries were classified as serious; nearly 50% of incidents occurred on streets or highways; and 34.1% of incidents occurred in school-zone areas. Head trauma was the most frequently injured region and a leading contributor to death. The key causes of incidents were failure to prioritise pedestrians, speeding, incorrect lane use, alcohol impairment, and unsafe following distances.ConclusionsThis study highlights child road traffic injury as a substantial public health burden in Ethiopia, with disproportionate impacts on school-aged children, males, adolescents, and those living in urban areas. These findings underscore the need for targeted public health strategies for child injury prevention. Enhancing community-level road safety education, strengthening enforcement of traffic laws, and promoting effective child supervision are essential for evidence-informed policy interventions. Implementation of these measures could meaningfully reduce child injuries and fatalities.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-026-27952-7.Keywords: Epidemiology, Traffic injuries, Children, Ethiopia, Road safety Background Road traffic injuries are a global public health issue that is steadily rising, resulting in substantial morbidity and mortality, especially among children [1, 2]. According to the 2023 World Health Organization (WHO) global status report on road safety, annual traffic fatalities have declined slightly to 1.19 million in recent years; however, the problem remains disproportionately high among children and adolescents. Indeed, road traffic injuries are the leading cause of death for children and adolescents between the ages of 5 and 19 years worldwide [3]. Approximately 95% of all child and adolescent deaths occur in low- and middle-income countries (LMICs) [4]. Children are particularly vulnerable road users because of their limited physical, cognitive and perceptual development. This developmental vulnerability limits their ability to accurately judge traffic speed, distance, and direction of vehicles [5]. Their smaller stature makes them less visible to drivers, while their relative physical fragility renders them more susceptible to serious injury than older children and adults. As a result, they face a higher risk of injury in unsafe traffic environments, leading to increased injuries and fatalities. Previous studies indicate that older children experience higher rates of injury. In contrast younger children, particularly those under five years of age, are at greater risk of death and long-term disability [6, 7]. Africa has the highest road traffic mortality rate globally, with approximately 20 deaths per 100,000 people in 2021, representing the highest fatality rate of any continent [8]. The African region experienced a 17% increase in road traffic deaths between 2010 and 2021 [9, 10], which accounts for nearly all road traffic fatalities, with Sub-Saharan Africa bearing the greatest burden. More than 600 children die on the roads every day, equivalent to one child death every two minutes, making children in LMICs reported to be eight times more likely to be injured than those in high-income countries [11]. Beyond fatalities, non-fatal injuries also have profound consequences for children, including physical, psychological, and developmental impacts [12, 13]. Early-life childhood injuries can cause long-term neurobiological and physical impairments that affect childhood growth and development [14]. Empirical evidence suggests that childhood injuries are linked to cognitive, emotional, and physical deficits that may extend into adulthood [15]. Ethiopia has one of the world’s highest burdens of road traffic injuries and fatalities, with mortality rates above the global average and a rising number of deaths, representing a severe road safety crisis [16]. While national Ethiopian fatality and injury rates are not consistently reported, the 2023 WHO country profile estimated approximately 28 road traffic deaths per 100,000 people [17]. Regarding data limitations, the absence of a comprehensive national injury surveillance system [18] has led to fragmented, under-reported and inconsistent data on the prevalence, patterns, and risk factors of road traffic injuries and fatalities in Ethiopia, including among children. Most existing studies in the country rely either on police crash data [19] or hospital-based data [20], which capture different subsets of cases resulting in considerable discrepancies between these sources, thereby limiting a complete understanding of the true burden of the problem. Reducing road traffic injuries and fatalities among children and adolescents has long been a global priority, highlighting their vulnerability and driving efforts to improve road safety. The WHO has declared a Second Decade of Action for Road Safety, with a target of a 50% reduction of road traffic deaths and serious injuries by 2030 [21]. To achieve this ambitious target, countries need to have robust, high-quality data to inform effective road safety interventions. In line with global road safety goals, Ethiopia has a national strategy aiming to reduce road traffic fatalities [22]. The Ethiopian road safety strategy planned to adopt the Safe System approach as a guiding framework to reduce injuries and fatalities. Most road traffic injuries are preventable through the comprehensive implementation of the five key pillars of the Safe System approach [23]. Because the strategies and principles of the Safe System approach were developed in high-income countries, its implementation faces barriers in many LMICs, such as Ethiopia, due to a lack of adequate research, inadequate funding, and poor data systems and weak road safety systems [24, 25]. Achieving this ambition may not be possible without context-specific research and evidence-based, coordinated injury prevention efforts. Investigating the epidemiological characteristics of these injuries is essential for several reasons: it supports the development of national child road safety policies, targeted prevention programmes, and child protection frameworks, and guides the prioritisation of evidence-based interventions. Despite the recognised public health problem of road traffic injuries in Ethiopia, the national prevalence, mortality, patterns, and associated risk factors among children and adolescents remain poorly understood. To our knowledge, this is the first systematic review and meta-analysis focusing on road traffic injuries among children in Ethiopia. This review aimed to consolidate and quantify existing evidence on road traffic injuries among the paediatric population, addressing the critical evidence gaps by examining the pooled prevalence, mortality, injury patterns, and incident characteristics. The findings will provide child-specific data to support the formulation of evidence-based policies and the design of strategic prevention initiatives to reduce injuries and fatalities. Methods Protocol and registration This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [26]. The study was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD420251073656). This systematic review and meta-analysis were guided by a clearly defined research question using the Population, Intervention, Comparison, and Outcome (PICO) framework. The PICO framework provided a systematic approach to identifying relevant studies, structuring the review, and interpreting the findings. For this review, the population comprised children under 18 years of age residing in Ethiopia. While a specific intervention was not directly applicable, road traffic injuries were treated as the primary exposure of interest. The comparison element was not applicable. The outcome of this study included prevalence, injury patterns, severity, mortality, and incident characteristics. Inclusion and exclusion criteria Studies were included if they met the following prior criteria: published in a peer-reviewed journal; in English, with full text available; involved children and adolescents in road traffic injuries between January 2000 and October 2025; and were quantitative studies examining epidemiological data on the burden, prevalence, patterns, or severity of road traffic injuries. Studies involving the general population were included only if they reported disaggregated data for children/adolescents. Studies were excluded if they met the following prior criteria: literature reviews (e.g., narrative or systematic), commentaries, case reports, proceedings, or dissertations; studies focusing on non-road transport-related injuries (e.g., air, water, or animal transport), as well as injuries due to falls, burns, or other causes; and studies limited to adult populations and those employing purely qualitative methodologies. Search strategy A comprehensive literature search was conducted to identify publications examining road traffic injuries among the paediatric population. A structured approach was designed to ensure the retrieval of all pertinent evidence related to road traffic injuries. The final search strategy was developed collaboratively by the research team and refined in consultation with an experienced academic librarian to ensure comprehensive coverage of all relevant literature. The search combined four groups of keywords: road traffic injuries (accidents, traffic, road traffic accident*, road traffic injur*, car accident*, motor vehicle accident*, car crash*, traffic-related injur*, traffic collision*, transport-related injur*, pedestrian injur*, cyclist injur*, bicycle injur*, vehicle occupant* injur*, motorcyclist injur*, road user* injur*, passenger* injur*.), children (child* or pediatric* or paediatric* or adolescen* or youth* or teen* or minor* or school-aged), epidemiology (prevalen*, inciden*, magnitude*, mortalit*, burden*, fatalit*, impact*, characteristic*, pattern*, epidemiolog*), and Ethiopia. Using appropriate Boolean operators (such as “OR” and “AND”) and Medical Subject Headings (MeSH) terms, the search strategy was applied across advanced database searches. The databases searched included Medline, Embase, CINAHL Complete, Ovid Transport, Scopus, and PubMed. Search terms such as ‘rider’, ‘moped’, and ‘e-scooters’ were omitted. We recognised that variations in terminology may influence the retrieval of relevant studies. However, we believe this impact is likely limited, as major concepts were captured through broader terms and supplementary searching techniques. In the Ethiopian context, motorcycles are the predominant two-wheeled mode transport, while mopeds and e-scooters are relatively uncommon and less frequently reported in the literature. Study screening and selection process Search results were imported into Covidence, a Cochrane-endorsed systematic review management platform. Two reviewers independently screened the titles and abstracts of all identified records against the predefined inclusion and exclusion criteria. Studies deemed potentially eligible or unclear based on the title and abstract were retrieved in full text for further assessment. Full-text screening was also conducted independently by two reviewers, with any discrepancies resolved through discussion or, where necessary, adjudication by a third reviewer. All screening decisions were recorded in Covidence. Studies that met all eligibility criteria were included in the final review. The overall screening and selection process was documented using a PRISMA flow diagram, which outlines the number of records identified, screened, excluded, and included, along with reasons for exclusion at each stage. Data extraction, recording, and management process Data extraction was conducted by two independent reviewers using a standardised data extraction template. The process commenced once potentially eligible studies were identified, and full-text articles were obtained through open access and uploaded to Covidence. Each reviewer independently extracted key information, and discrepancies were resolved through discussion to ensure accuracy and consistency. A structured data extraction template was developed in Microsoft Excel to ensure completeness and standardisation across studies. Fields included study identifiers: author(s), year of publication, study design, and study location or region; population characteristics: age range of participants, sample size, gender distribution, and type of road user (e.g., pedestrian, passenger, or cyclist); outcome measures: prevalence and patterns of road traffic injuries and types and severity of injuries; data sources: setting in which the data were collected (e.g., hospital-based, community-based, or police records); and main findings: results directly relevant to the outcomes of interest and results of quality assessment. To ensure reliability, approximately 10% of the included studies were randomly selected and cross-checked by both reviewers. No discrepancies were identified. Quality assessment To ensure the validity and reliability of our findings, we systematically reviewed and evaluated the risk of bias and the methodological quality of all included studies. The methodological quality of included studies was assessed using standardised critical appraisal tools developed by the Joanna Briggs Institute (JBI) [27]. The two reviewers independently rate the quality of each article based on the criteria of the JBI assessment tool. Each study was evaluated according to this tool. The possible responses were “yes” if the criterion was clearly met, “no” if the criterion was clearly not met, “unclear” if insufficient information was available to determine whether the criterion was met or not met, and “not applicable” if the criterion did not apply to the specific study. The scoring method was 1 point for ‘yes’ responses and 0 points for ‘no’, ‘unclear’, and ‘not applicable’ responses. The two independent reviewers reduce subjective bias and ensure consistency. Finally, the quality of each study was classified as good, fair, or poor, corresponding to scores above 70%, between 50% and 70%, and below 50%, respectively. Data analysis The extracted data were exported from a Microsoft Excel spreadsheet to Stata software (version 18) for statistical analysis. The details of the included studies were systematically reviewed and presented in tables. The epidemiological characteristics of children, including age and gender categories, road user type, injury severity, incident location, and vehicles involved, were described and presented in text and tables. Random effects models were used to estimate the pooled prevalence of injuries and deaths. Summary effect estimates with 95% confidence intervals (CIs) were presented using a forest plot. Heterogeneity between the included studies was assessed using the index of heterogeneity (I2) and Cochran’s Q statistic with the corresponding 95% CI. Significant heterogeneity was considered when I2 was greater than 50%, and the p-value was less than 0.05. Funnel plots and Egger’s test statistics were conducted to evaluate the presence of publication bias. To explore the source of heterogeneity, subgroup analysis was conducted using study characteristics as moderators, including sample size, year of publication, and geographical region. Additionally, sensitivity analysis was performed to evaluate the stability and reliability of the pooled effect estimates. Results Search results The systematic search yielded 5774 records. Of these, 3,504 records were retained after removing 2,270 duplicates. After the initial title and abstract screening, 3,361 records were excluded, leaving 143 records eligible for further full-text screening. After the full-text screening, 105 articles were excluded due to the following reasons: no full text available (n = 27), not conducted in Ethiopia (n = 19), focuses only on adults (age ≥ 18 years) (n = 13), published outside date range (n = 1), not peer-reviewed (n = 3), not original research (n = 6), injuries unrelated to road transport (n = 30), and not a full research article (n = 5). Finally, 36 studies were included in this systematic review and meta-analysis (Fig. 1). PRISMA flow diagram of the study selection for systematic review and meta-analysis Study characteristics The 36 peer-reviewed articles involved approximately 55,397 participants across all regions of Ethiopia. The sample sizes of the included studies ranged from 123 [16] to 25,237 [28]. Ten studies [28–36] were conducted among children only, and twenty-six studies [16, 20, 37–61] were conducted among both children and adults. Twenty studies were registry-based, using secondary data from hospital records. Five studies were interview-based with road traffic injury victims and their caregivers; five studies were surveys among road traffic injury victims and their families; four studies used data from traffic police records; two studies used data linkage from both hospital and police data sources using the capture-recapture method; and one study used data from the Global Health Estimates. The included studies were conducted in various regions across the country. The largest number of studies was conducted in Addis Ababa (the capital city of Ethiopia) with 11 studies [32, 36, 38, 39, 43, 44, 49, 51, 53, 54, 57]. Seven studies [31, 42, 46–48, 56] were conducted in the Oromia region; five studies [20, 29, 30, 33, 41, 61] were conducted in the Amhara region; five studies [35, 40, 45, 52, 58] were conducted in the Sidama region; two studies [16, 28] were nationwide; one study was conducted in Dire Dawa city [37]; three studies [34, 50, 55] were conducted in the Tigray region; and two studies [59, 60] were conducted in the Southern Nations, Nationalities and Peoples’ Region. The distributions of injuries by road user types, injury severity, and prevalence of child road traffic injuries and fatalities in each included study were summarised in the table (Table 1). In addition, the study’s key findings, proposed intervention strategies, and overall quality were assessed using the JBI quality assessment tool (Additional file 1). Table 1Characteristics of included studies and distribution of RTIs by road user type, injury severity, and prevalence of child RTIs and fatalitiesAuthorsData sourcePercentage of RTIs by road user type (%)Severity of RTIs (%)Child RTIs (%)Child fatality (%)Taye et al. 2025 [29]Hospital recordsPedestrian (55.5), vehicle occupant (26.1), motorcyclist (18.4%)NR10010.7Kebede et al. 2023 [19]Hospital recordsNRNR24.218.37Degu et al. 2024 [20]Hospital recordsNROverall: serious (39.5), Moderate (37.6), minor (22.8)32.133.9Berhe et al. 2014 [36]Police recordsPedestrian (95.7), passengers (2.3), bicycle (0.2)Slight (37.6), severe (50.0), fatal (12.3)100.012.3Abegaz et al. 2019 [16]EDHS 2016Overall: (Passengers(35.0), drivers (21.9), motorcyclists (23.0), pedestrian (12)NR11.4NRTulu et al. 2025 [37]Police and hospital recordsOverall: Pedestrian:34.1, Driver: 39.9, passenger: 14.5Motorcyclist:3.5, bicyclist: 7.8Children < 18: serious: 37.8 from hospital27.727.0Mulugeta et al. 2024 [38]Hospital recordsNRNR29.781.7Deguale, 2024 [39]Hospital recordsOverall: Pedestrian (51.0), passengers (27.0), drivers (22.0)NR18.3NRMolla et al. 2024 [30]Hospital recordsPedestrian (94.3), passenger (5.7)NR13.10.7Afacho et al. 2024 [40]Hospital recordsOverall: pedestrian (35.18), passenger (28.39), driver (36.43)NR8.0NRDeresse et al. 2023 [41]Police recordsOverall: passenger (52.7), pedestrian (3.2), drivers (44.1)NR6.3 (drivers < 18)3.8 (drivers under 18)Mamo et al. 2023 [42]Hospital recordsOverall: pedestrian (49.6), passenger (26.1), driver (23.5), assistant driver (0.8)Overall: severe (49), moderate (37.0), minor (14.3)21.0NRBezabih et al. 2022 [43]Hospital recordsOverall: Pedestrian (51.8), Passenger:36.3others (1.2) Bicycle:8.3NR13.7 (children < 13)NRDagnaw et al. 2022 [31]Hospital recordsPedestrian (58.2), passenger (25.5), cyclists (16.3)Serious (35.8), moderate (56.0), minor (8.2)24.212.2Hailu et al. 2022 [44]Hospital recordsOverall: Passenger (23.9), pedestrian (14.3), motorcyclist (2.8), cyclist (0.5)NRNRNRAklilu Toma et al. 2021 [45, 62]Police recordsOverall: pedestrian (20.14), other types were not reportedNR38.4NRDemisse et al. 2021 [46]InterviewOverall: passengers (53.8), pedestrian (32.3), driver and motorcyclists (13.9)NR27.8 (age 15–24)14.3Woyessa et al. 2021 [47]InterviewOverall: pedestrian (25.5), passengers (40.9), driver (19.6), driver assistant (14.0)Overall: mild (7.81), moderate (68.03), severe (24.16)13.6NR, overall (38.8%)Deresse et al. 2021 [48]Hospital recordsOverall: pedestrian (44.4), passenger (41.4), others (14.2)Overall: mild (42.4), moderate (34.9), severe (22.7)21.1NRGebresenbet et al. 2019 [49]Hospital recordsOverall: pedestrian (25.8), passenger (54.6), driver (19.6)Overall: severe (36.6), non-severe (63.4)12.2NR Tesfay et al. 2019 [50]Hospital recordsOverall: passenger (44.7), pedestrian (37.9), driver (16.5), other (0.9)Overall: moderate (82.0), severe (18.0)13.3 (age<14)NRBacha et al. 2020 [21]Hospital recordsPedestrian (78.6), passenger (18.2), driver (3.2)NR39.8NRBaru et al. 2019 [51]InterviewOverall: pedestrian (39.7), driver (10.7), vehicle occupant (38.8), motorcyclist (10.7)Overall: severe (36.4), not severe (63.6)20.4NRTiruneh et al. 2019 [33]Interview administeredPedestrian (77.1), passenger (22.9)NR20.0NRLi et al. 2018 [28]GHE 2015NRNRNR25.0Hagos, 2017 [34]Hospital recordsNRNR24.84.9Negussie et al. 2018[52]Hospital recordsNROverall: minor (56.7), moderate (25.7), severe (9.3)NRNRTiruneh et al. 2014 [54]InterviewOverall: pedestrian (71.7), passenger (13.0), drivers (12.2), driver assistant (3.1)Overall: minor (7.6), moderate (74.4), severe (15.0)16.3NRAmdeslasie et al. 2016 [55]Hospital recordsOverall: pedestrian (17.6), vehicle occupant (61.2), not specified (16.5), other (4.7)NR20.8NRAbegaz et al. 2014 [56]Police and hospital recordsOverall: vehicle occupant (66.9), pedestrian (20.3), bicycle/motorcycle (6.12)NRNR43.4Bedry et al. 2020 [35]Hospital recordsPedestrian (83.3), vehicle occupant (16.7)NR45.4NRAlemayehu et al. 2023 [57]Police recordsOverall: pedestrian (80.0), passenger (12.0), driver (5.0), other (3.0)NR22.48.0Duko et al. 2019 [58]Hospital recordsOverall: passenger (44.0), driver (31.5), pedestrian (24.5)NR12.0NRFite et al. 2020 [59]Hospital recordsNRminor (54.8), moderate (33.3), severe (12.6)NRNRHailemichael et al. 2015 [60]Hospital recordsOverall: pedestrian (33.8), passenger (45.8), driver (20.5)NR32.5NREDHS Ethiopian Demographic and Health Survey, GHE Global Health Estimate, NR Not reported, RTIs Road traffic injuriesIdentification of eligible studies from major databases and citation searching Characteristics of included studies and distribution of RTIs by road user type, injury severity, and prevalence of child RTIs and fatalities Overall: (Passengers (35.0), drivers (21.9), motorcyclists (23.0), pedestrian (12) Overall: Pedestrian:34.1, Driver: 39.9, passenger: 14.5 Motorcyclist:3.5, bicyclist: 7.8 Overall: Pedestrian (51.8), Passenger:36.3 others (1.2) Bicycle:8.3 EDHS Ethiopian Demographic and Health Survey, GHE Global Health Estimate, NR Not reported, RTIs Road traffic injuries Identification of eligible studies from major databases and citation searching Epidemiologic characteristics of child road traffic injuries Gender distribution All included studies reported that male children were overrepresented among road traffic injury victims compared with females. The proportion of road traffic injuries among males ranged from 52.1% [47] to 91.8% [45]. Most studies reported that the proportion of males affected by road traffic injuries exceeded 60%, with an average of 71.24%. Age distribution Six studies reported the mean age of children; the combined mean age of children was 8.82 ± 4.25 years. Four studies [34–36, 61] reported that school-aged children (5–12 years) were the over-represented age group for road traffic injuries. Similarly, four studies [29, 31, 32, 54] identified adolescents (13–18 years) as the most affected group. The overall estimates of age distribution from eight studies indicated that school-aged children (aged 5 to 12 years) and adolescents (aged 13 to 18 years) were the most affected age groups by road traffic injuries, accounting for 36.26% and 40.63%, respectively. Road traffic injuries by residential area The majority of studies reported that residents in urban areas were more affected by road traffic injuries than those from rural areas, with proportions ranging from 54.8% [48] to 83.7% [55]. Only five studies reported that the majority of the affected population were from rural areas, ranging from 53.7% [59] to 69.6% [16]. Overall, most studies reported that road traffic injuries were more common among urban residents. Patterns and severity of child road traffic injuries Head and limb injuries were the most commonly injured body regions following road traffic injuries. Several studies have examined the head as a common injured body part. In addition, the upper and lower limbs were also commonly injured body parts. Other injured body parts are the abdomen, neck, face, chest, spinal cord, soft tissue, and multiple injuries. Several studies [36, 42, 45, 57, 58] proved that the causes of crashes were mainly related to driver errors, such as failure to prioritise pedestrians, high speed, carelessness, failure to prioritise other vehicles, driving in the incorrect lane, alcohol impairment, and failure to maintain a safe distance between vehicles. These factors were synthesised narratively. This is because only three studies [36, 58, 62] reported these variables in sufficient detail, given this limited number of studies, a pooled quantitative synthesis was not appropriate. Eight studies [20, 31, 34, 36, 37, 49, 51, 54] reported the severity of road traffic injuries; the overall estimated proportion of serious injuries was approximately 37%. Five studies [20, 31, 36, 37, 54] reported minor injuries with an overall weighted proportion of 36.46%. Locations of child road traffic incidents The studies highlighted the predominant locations of child road traffic incidents [31, 32, 34–36]. The findings consistently identified three primary locations where these incidents were most prevalent: streets or highways, school zones, and residential areas. Among these, streets/highways were the most frequent sites of child road traffic incidents, accounting for about half (49.3%) of all reported injury cases. School zones were the second most common location for incidents, with 34.1% of injuries occurring around them. Lastly, incidents in residential areas accounted for 16.7% of the incidents, indicating that these areas are also risky locations for child road traffic incidents. Collision types in child road traffic incidents Several studies [20, 30–33, 36, 46, 58, 60] clearly identified the types of vehicles involved in collisions that resulted in child road traffic injuries. The weighted proportions of the vehicles involved in the incidents were analysed. Cars (four-wheeled vehicles such as automobiles, taxis, and buses) accounted for 59.7% of all injury cases, followed by motorcycle incidents (21.5%) and bajaj incidents (11.3%), a three-wheeled vehicle commonly used in Ethiopia. The remaining 7.5% of child road traffic injury cases accounted for bicycle incidents. Road user patterns in child injuries Seven studies [29–33, 35, 36] analysed the pattern of child road traffic injuries by road user type. The sample size and the proportion of road users who sustained injuries were used to analyse the pooled percentage of injured road users. The overall weighted proportion of road traffic injury victims by road user type was analysed, and pedestrians were found to be the most affected group, accounting for approximately 81.5% of all reported road traffic injuries. Passengers were the second-most affected group, comprising 15.1% of cases. Child motorcyclists (2.7%), cyclists (1.1%), and drivers under 18 years of age (0.2%) were considerably less common. Meta-analysis Prevalence of child road traffic injuries In this systematic review and meta-analysis, we investigated the pooled prevalence of road traffic injuries among the paediatric population. In this pooled estimate, we included only studies that provided clearly disaggregated data for children. When such distinctions were unclear, we carefully assessed the study context and excluded those studies from the meta-analysis to minimise potential bias. Of the 36 studies, 27 clearly reported a child-specific proportion of road traffic injuries. The pooled prevalence of road traffic injuries was estimated at 23% of all trauma cases with a 95% CI (21%-25%). Notably, there was heterogeneity among the studies (I2 = 96.14%, p < 0.001), prompting a random-effects model analysis. The pooled estimates, given a high index of heterogeneity, should be interpreted with caution. Differences in study design, population characteristics, and case ascertainment may have influenced the observed prevalence. The random-effects model indicated that the precision of the pooled prevalence of road traffic injuries was high, as evidenced by a narrower confidence interval (Fig. 2). Pooled prevalence of road traffic injuries among children with traumatic injuries in Ethiopia Publication bias The presence of publication bias was assessed for the pooled prevalence of road traffic injuries using graphical funnel plot and statistical evaluation with Egger’s test. The funnel plots visualise the potential publication bias by showing study precision against effect size. At the same time, Egger’s test was used to assess publication bias based on the statistical value (p-value). The funnel plot appears slightly asymmetric, suggesting possible small study effects. The asymmetry or missing dots on one side suggests that studies reporting smaller or negative effects might not have been published or might have been underreported (Fig. 3). The confirmatory statistical test using Egger’s test indicated that there was no statistically significant evidence of publication bias or small study effect (p = 0.1279). The Egger’s tests generally provide more accurate statistical evidence and are preferred over funnel plots (a subjective visual interpretation). Funnel plot assessing publication bias in the pooled prevalence of child road traffic injuries Regional distribution of child road traffic injuries Analysing the regional distribution of road traffic injuries helps identify hotspot areas for targeted interventions and resource allocation. We analysed the regional prevalence of road traffic injuries among children in Ethiopia. The prevalence of road traffic injuries among children remains high across all regions. The prevalence of road traffic injuries varied notably between the regions. The highest prevalence was observed in the Southern Nations, Nationalities, and Peoples’ Region (SNNPR) at 33% (95% CI: 28%-37%), followed by Dire Dawa City at 28% (95% CI: 26%-30%), the Tigray region at 25% (95% CI: 12%-37%), the Sidama region at 26% (95% CI: 8%-44%), the Amhara region at 22% (95% CI: 11%-32%), the Oromia region at 22% (95% CI: 17%-27%), and Addis Ababa at 22% (95% CI: 15%-29%). The test for subgroup differences showed a statistically significant variation in the prevalence of road traffic injuries across the regions (Q = 14.68, p = 0.04) (Fig. 4). Forest plot illustrating the regional variation in the prevalence of road traffic injuries among children in Ethiopia Prevalence of mortality among child road traffic victims Of the included studies, 13 clearly reported the proportion of child mortality attributable to road traffic injuries. We used the meta-analysis to estimate the overall mortality of children following road traffic injuries. The result of the random-effects model indicated that the overall pooled prevalence of mortality among children with road traffic injuries was 17% with a 95% CI (11%-24%). The statistical values indicated significant heterogeneity among the studies (I2 = 99.17%, p-value < 0.001). A single summary of pooled estimates may mask important variabilities in study characteristics. Therefore, it should be interpreted cautiously (Fig. 5). In addition, the included studies used to estimate the pooled prevalence of road traffic injury mortality were assessed for publication bias. The visual analysis of the funnel plot indicates slight asymmetry. The most sensitive statistical test (Egger’s test) was used to assess publication bias or small study effects in the pooled mortality of children from road traffic injuries. Egger’s statistical test indicated no significant evidence of publication bias (p = 0.5645) (Fig. 6). Pooled prevalence of mortality among children with road traffic injuries in Ethiopia Funnel plot illustrating potential publication bias in the pooled mortality estimates of child road traffic injuries Subgroup analysis Since the meta-analysis demonstrated apparent heterogeneity between the included studies, a subgroup analysis was conducted using categorical variables such as sample size, study region, study design, and year of publication to identify the source of heterogeneity. The subgroup analysis, based on the sample size, showed no heterogeneity in the overall statistical analysis (I² = 0%, p = 0.61). Similarly, the subgroup analysis based on study design and year of publication indicated no significant heterogeneity between the studies, with p-values of 0.713 and 0.948, respectively. Subgroup analysis by data source indicated some variability in the pooled prevalence estimates. According to the Cochrane Handbook, there was moderate heterogeneity among the subgroup survey data sources (I2 = 39.7%). However, the overall test for subgroup differences was not statistically significant (Qb = 3.06, p = 0.802), indicating that the data source did not significantly influence the pooled estimates. However, subgroup analysis by region of study showed statistically significant heterogeneity across the studies. Regions were a significant source of heterogeneity (overall I² = 98.42% and p-value = 0.04), and heterogeneity within regions was also high. Additionally, heterogeneity may be explained by other variables not included in this review. Sensitivity analysis Sensitivity analysis was performed by removing one study at a time (a leave-one-out analysis) to assess the influence of individual studies on the overall pooled prevalence estimate. This involved sequentially excluding individual studies to assess the influence of each on the overall pooled estimate. The results indicated that removing a single study from the analysis did not significantly influence the pooled prevalence estimate of child road traffic injuries. Therefore, the pooled effect estimate remained stable when any single study was excluded. Discussion This systematic review and meta-analysis investigated the epidemiological evidence, injury patterns, and incident characteristics among children. The review clearly highlighted that road traffic injury was a major public health problem among children and adolescents in Ethiopia. The study assessed the epidemiological characteristics of road traffic injuries, including prevalence, mortalities, and injury patterns by age and gender, residential and geographical regions, accident locations, and road user type. The current study revealed that the pooled prevalence of road traffic injuries among children accounted for 23% of all trauma-related injuries. Estimating the pooled prevalence from both child-only studies and general population studies reporting child-specific data is methodologically acceptable in epidemiological synthesis because it maximises data availability and improves the representativeness of the pooled estimates. In addition, it allows a more comprehensive assessment of the available evidence. This is particularly true in settings such as Ethiopia, where child-focused studies are limited. This prevalence was notably higher than that reported in similar studies conducted in Mozambique (14.3%) [63] and in India (17%) [64]. This might be due to several reasons such as unsafe road user behaviour, inadequate road safety infrastructure, poor public awareness, limited traffic regulation and enforcement in Ethiopia [65, 66]. However, the prevalence of child road traffic injuries in Ethiopia was relatively comparable to that reported in urban areas of South Asia and Saudi Arabia, where the prevalence was reported to be 22% [67] and 25% [68], respectively. In Ethiopia, poor road safety infrastructures are among the contributors to the increased prevalence of child road traffic injuries. Many school zones lack proper pedestrian crossings, which puts children at risk of road traffic crashes. Urban areas with vehicle-pedestrian mixed traffic were identified as high-risk zones for child injuries, highlighting the need for stronger road safety measures in high-traffic areas. Additionally, hazardous traffic conditions, such as speeding vehicles, unsafe crossings, and poorly designed pedestrian zones, may further increase child road traffic injuries [69]. There was a substantial fatal outcome following road traffic crashes involving children. The mortality among victims of child road traffic injuries was estimated at 17%. This finding was in line with other studies in Nigeria (18%) [70], Ghana (19%) [71], and Malaysia (15.5%) [72]. However, mortality was higher than in other studies in Northwest India (9.4%) [73], India (8%), Turkey (7.4%) [74], and Romania (5.8%) [75]. This difference might be associated with several road safety challenges in Ethiopia, such as poor road infrastructure, weak law enforcement, inadequate vehicle safety features, a large proportion of vulnerable road users, and rapid motorisation [76–78]. Furthermore, post-crash emergency medical services are limited by a lack of trained personnel, poorly equipped ambulances, and weak coordination systems. The pre-hospital care for crash victims is limited, and delays in reaching hospitals are common in Ethiopia, particularly in rural areas. As a result, many road traffic injuries lead to preventable fatal outcomes [79]. In the current study, males were more than twice as likely to be involved in road traffic injuries as compared to their female counterparts. Male children accounted for an average of 71% of all injury cases. This finding was in line with other studies conducted in Lebanon (73%) [80] and Romania (78.5%) [81]. This finding was also consistent with various injury surveillance studies conducted in Sub-Saharan Africa [82], Iran [83], and Finland [84]. The higher prevalence of road traffic injuries among boys may be due to greater exposure to traffic environments and higher risk-taking behaviour. Compared to females, males are more frequently involved in outdoor and road-related activities, including cycling and motorbike riding, often without protective equipment [85]. In addition, boys are less likely to use safety measures such as helmets and seat belts and tend to engage in more running and active play, all of which contribute to their higher injury rates. In contrast, girls are more likely to walk with companions, making boys more vulnerable to injuries [86]. The results of this review showed that the majority of child road traffic injuries occurred among pedestrians. Similar findings were reported by other studies [87, 88]. This might be because walking is a common mode of transport for children and adolescents in Ethiopia; they often travel long distances to school, market areas, and roads without adequate supervision, sidewalks, or traffic signals. Limited public awareness of road safety and inadequate urban planning further increase the risk, making roads especially dangerous for child pedestrians in both urban and rural settings of Ethiopia [88, 89]. Our review found that school-aged children aged 5 to 12 years (36.26%) and adolescents aged 13 to 18 years (40.65%) were more frequently involved in traffic incidents than younger children. This finding was supported by other studies [90, 91]. Increasing age is associated with greater independence and mobility among older children and adolescents, as they are more likely to play near roads, walk to school, or use bicycles and other forms of transport without adult supervision. School-aged children may engage in risky behaviours, such as running across the street or crossing the road without properly assessing the traffic environment [92]. This study found that head and limb injuries were the most common types of injuries sustained by children and adolescents in road traffic crashes. This finding aligns with other studies conducted in different countries, such as South Korea [93], Saudi Arabia [94], and Turkey [95]. Similarly, research findings from Europe [96] reported high rates of head and extremity injuries among paediatric trauma patients. These consistent patterns across different settings highlight the universal vulnerability of children’s heads and limbs in traffic incidents that need targeted safety measures such as child restraints, helmet use, and pedestrian protections. To prevent frequent head injuries and associated fatalities, helmet use should be strictly enforced. However, there is no law enforcement for helmet use in Ethiopia, and helmet-wearing rates are low [97]. The current study found that the most common locations for child road traffic crashes in Ethiopia were streets/highways and school zones, accounting for 49.3% and 34.1%, respectively. These areas were identified as high-risk zones for child road injuries that require attention. The justification is that these environments pose a high risk due to unsafe traffic manoeuvres, including high or inappropriate speed; lack of pedestrian infrastructure; and limited child supervision, especially during school hours and in densely populated areas. This finding was supported by other studies [98, 99]. Another study reported that fatal child pedestrian crashes were highest in school zones [100]. This review indicated a residential variation in child road traffic injuries, with most studies reporting a higher prevalence of injuries within urban areas. This finding is supported by other studies [101, 102]. Higher traffic density and increased exposure to vehicles in urban areas may contribute to a higher prevalence of injuries among children travelling there. In contrast, a few studies found that rural children were more affected by road traffic injuries. Similar studies conducted in South Africa [103] and China [104] reported a higher prevalence of road traffic injury among children from rural areas. Studies conducted in rural hospitals and local areas, where rural populations were over-represented, may report higher injury rates among rural children. Limitations of the study This study investigated the epidemiological evidence of road traffic injuries among children. Despite yielding significant findings, acknowledging the study’s limitations remains essential. Firstly, child-specific studies were limited in Ethiopia. To improve representativeness, we also included general population studies that reported separate child-specific data. Although only studies with clearly disaggregated estimates for children were included, differences in study design, sampling frames, and population characteristics between these study types may have introduced additional variability and potential bias in the pooled estimates. This should be considered when interpreting the findings. Secondly, this review demonstrated a substantial heterogeneity across studies. While high heterogeneity is common in meta-analyses of prevalence data due to differences in study populations, measurement approaches, and settings, it nonetheless limits the interpretability and generalisability of pooled estimates. Although subgroup and sensitivity analyses were conducted, much of the variability remained unexplained, suggesting that additional unmeasured factors may contribute to the observed differences. Accordingly, the pooled estimates should be interpreted with caution. Another issue concerns the search terms. Although a comprehensive search strategy was used, it is possible that some studies may not have been identified due to variations in terminology (e.g., alternative descriptors for road users or crash outcomes). This may have resulted in the omission of a small number of relevant studies. Finally, because the included studies were descriptive, this review did not investigate the multi-level risk factors associated with child road traffic injuries. Future research should focus on a comprehensive risk factor analysis using a well-designed analytical approach. Implications and conclusion Overall, this review identified several key findings applicable to the implementation of road traffic injury prevention among the paediatric population in Ethiopia. The review provides epidemiological evidence and incident characteristics on child road traffic injuries and outlines implications for enhancing child road safety. The review found that crashes were mainly caused by driver errors, such as failure to give priority to pedestrians, high speed, failure to give priority to other vehicles, driving in the incorrect lane, drinking alcohol, and failure to maintain distance between vehicles. This finding provides direct support for policymakers to strengthen enforcement of traffic laws related to driver behaviour. This study found that streets and highways were the most common sites of road traffic crashes, reinforcing the need to strengthen traffic control mechanisms to facilitate safe traffic flow on streets and highways. School zones were also identified as high-risk areas for child injuries; therefore, proper enforcement and improvement to road infrastructure are required, including the installation of speed bumps, visible signage, and crossing guards to protect children, particularly during school hours. Implementing school crossing supervisors is strongly recommended to keep children safe and reduce pedestrian injuries, particularly during school travel times. In addition, residential areas were also high-risk zones for child traffic injuries, which require attention. Furthermore, head injuries were the most commonly injured body part; this indicates the need to reinforce appropriate safety measures such as strict helmet use for cyclists and motorcyclists. The current findings indicated that pedestrians were the most affected road users in road traffic injuries, highlighting the need to provide education and training for children and their parents on safe pedestrian behaviours and traffic safety. Parents and caregivers of children can also benefit from road safety education to better supervise their children in road traffic. School-aged children and adolescents were identified as the age group most affected by road traffic injuries. Stakeholders, such as schools and road transport officials, should provide appropriate road safety education for children and adolescents as well as providing transport options other than walking (e.g., school buses). In addition, this study demonstrated regional variation in the prevalence of child road traffic injuries across the country. This regional variation could help policymakers allocate resources to contextual road safety interventions. Because pedestrians are disproportionately affected, streets and highways are recognised as high-risk environments; implementing the Safe System approach may prove effective in preventing such injuries by fostering safer road users, safer road infrastructure, safer vehicles, and post-crash response. Importantly, the Safe System approach acknowledges that human error is inevitable and therefore advocates for designing systems that prevent such errors from resulting in fatalities or serious injuries. The findings of this study will offer valuable insights for stakeholders involved in injury prevention and serve as input for planning road safety interventions targeted at these high-risk populations, particularly in LMICs. Supplementary Information Supplementary Material 1. Acknowledgements Not Applicable. Abbreviations Confidence Interval Joanna Brigs Institute Low- and Middle-Income Countries Population Intervention Comparison Outcome Preferred Reporting Items for Systematic Reviews and Meta-Analyses International Prospective Register of Systematic Reviews United Nations Children's Fund World Health Organization Authors’ contributions MSA: Conceptualisation, Methodology, Software, Data curation, Formal analysis, investigation, Writing- Original draft preparation **, AD: Supervision, Methodology, Visualisation, Data curation, Writing- Reviewing and Editing. **JO**: Supervision, Visualisation, Methodology, Investigation, Writing-Review and Editing. **SK**: Supervision, Methodology, Validation, Resources, Project administration, Investigation, funding acquisition, Writing- Reviewing and Editing. All authors read and approved the final manuscript. Funding No funding was accepted for conducting the study and preparation of this manuscript. Data availability All data are available upon request. The reader could contact the corresponding author. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement All data are available upon request. The reader could contact the corresponding author.