Prevalence and risk factors for road traffic accidents and mortalities in the Amhara region, Ethiopia: a systematic review and meta-analysis

Abstract

Introduction

Road traffic accidents are a serious public health challenge in Ethiopia’s Amhara region. This study provides pooled estimates of road traffic accidents related injuries, death and risk factors.

Methods

Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we systematically searched PubMed/Medline, Web of Science, Scopus, Google Scholar, African Journals Online, EMBASE databases and grey literatures from 10 January to 15 March 2025 using Medical Subject Headings terms and Boolean operators. Studies published since 2020 were screened to ensure the inclusion of the most recent and contextually relevant evidence, and data were extracted independently by two reviewers using a standardised Excel sheet form. Quality was assessed via the Joanna Briggs Institute checklist. Random-effect models were employed due to heterogeneity. Meta-analysis was conducted using STATA V.14.0 software. Subgroup analysis and meta-regression were performed to identify factors and source of heterogeneity. Small study effects and publication bias were examined through funnel plots and Egger’s/Begg’s tests.

Results

12 studies (n=5222) were included. Pooled prevalence of road traffic accidents related death was 33% (95% CI 23% to 48%) and injury 48% (95% CI 28% to 58%). Meta-regression showed no significant effects of publication year, zone, study design or sample size. Key risk factors included driver-related issues (young age, inexperience, speeding, phone use, alcohol/fatigue driving, overloading), poor vehicle/road and environmental conditions.

Conclusions

Road traffic accidents related injuries and death burden remain high in the Amhara region, driven mainly by risky driver’s behaviour, suboptimal infrastructure and environmental conditions. Strengthening safety regulations and sustained behavioural interventions are urgently needed.

PROSPERO registration number

CRD 420251003488

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Road traffic accidents (RTAs) are a major public health problem in the Amhara region. Driver behaviour, vehicle conditions and road infrastructure are key contributing factors.

WHAT THIS STUDY ADDS

• This study provides the first pooled estimates of RTA-related injuries (48%) and deaths (33%).

• It identifies driver behaviours and vehicle/road conditions as leading risk factors, based on evidence synthesised from multiple studies.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Given the high burden of RTAs, ongoing efforts are needed through stricter law enforcement, expanded driver education, targeted behavioural interventions and improved safety regulations and infrastructure.

Introduction

Road traffic accidents (RTAs) are a serious public health challenge globally, which is responsible for approximately 1.3 million deaths annually and injuries to 20–50 million people.¹ It also has a multifaceted impact on the socioeconomic aspects.¹ Of these, over 90% of road traffic fatalities occurred in low and middle-income countries.² Africa accounts for a large share of 26.6 deaths per 100 000 people; than the global average of 17.5 per 100 000 people this translates to roughly 675 people dying on African roads every day.³

The problem also arises in Ethiopia with a significant magnitude. According to the Federal Police Commission in the 2021 report, 15 034 road accidents resulting in 4161 deaths and 10 873 injuries occurred.⁴ Despite efforts to address road safety issues, Ethiopia continues to struggle with rising rates of RTAs and is placed at the highest RTA rates in Africa, with nearly 73 traffic deaths per 100 000 people usually.⁵ A systematic review in Ethiopia shows that the pooled prevalence of RTAs is 36%.⁶

The Amhara region, in particular, faces a high burden of RTAs, contributing significantly to the national statistics. These accidents not only result in loss of life but also impose substantial health, economic and social consequences. The economic impact includes direct costs such as medical expenses and indirect costs like lost productivity, which can strain local economies and healthcare systems. Furthermore, RTAs can lead to long-term disabilities, affecting family dynamics and social stability.⁷

The prevalence of RTAs in the Amhara region varies, with studies indicating a range of factors contributing to these accidents. It occurs due to various factors that are induced by human factors, such as driver licences,^{8 9} age of driver,^9–11 farmers and students,⁹ driver experience and driver education level,¹⁰ failure to give priority to pedestrians,^{11 12} distances between vehicles,¹² sex of driver, alcohol consumption, overloading and overspeed,¹¹ vehicle conditions like vehicle service year, poor maintenance and inadequate safety features,^{10 13} environmental factors such as location of road (urban or rural), and daylight hour,⁹ season of accident,¹⁰ days of week^{10 11} and road infrastructure factors like poor access to traffic lights and road condition.^{11 14}

RTAs are an urgent health issue in the Amhara region, resulting in severe morbidity, mortality and economic difficulties. The region has an extensive road network that connects major interregional trade and transport corridors, resulting in high traffic volumes. This network comprises a mix of urban, peri-urban and rural roads with varying safety standards. The predominance of long-distance passenger and freight transport, often involving minibuses and heavy trucks, has contributed to one of the highest RTAs, related injuries and fatalities in Ethiopia. However, there is limited aggregated data on the prevalence and associated factors of RTAs in the Amhara region, highlighting the need for a systematic review and meta-analysis.

The contribution of this research lies in its focus on filling a critical gap in understanding the prevalence of RTAs and its associated factors for morbidity and mortality in Amhara region, providing a foundation for evidence-driven intervention measures. These insights will equip policymakers and road safety authorities with the evidence needed to design targeted interventions in the region. By implementing evidence-based policies aimed at identifying main risk factors associated with RTAs, prevalence and severity of RTAs in the Amhara region can be reduced.

Materials and methods

Study protocol and registration

This systematic review and meta-analysis was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines¹⁵ (online supplemental checklist S1) to ensure transparent and comprehensive reporting. The study protocol (online supplemental file S1) was developed a priori and registered with PROSPERO under the registration number CRD420251003488.

Study design and search approach

A systematic review and meta-analysis of published articles were conducted in the Amhara region of Ethiopia, from 10 January 2025 to 25 March 2025, the period during which we conducted the literature search and data collection for this review, not the timeframe of the data reported in the included studies. The search strategy involved a comprehensive consideration of major databases including PubMed/Medline, Web of Science, Scopus, Google Scholar, African Journals Online, EMBASE databases and grey literature such as theses and dissertations. The latest evidence and recent publications were sourced through direct engagement with experts and researchers. Medical Subject Headings terms related to RTAs, magnitude or prevalence, associated factors or risk factors or determinants or causes, Amhara region, and studies published since 2020 were used during the search process (online supplemental table S1). The study period beginning in 2020 ensures the inclusion of the most recent and contextually relevant evidence. This timeframe captures data reflecting contemporary traffic conditions, enforcement practices and behavioural trends, while minimising the influence of outdated infrastructure or policy contexts. Moreover, post-2020 studies benefit from improved data collection and reporting systems, thereby enhancing the reliability and applicability of findings for current policy formulation and intervention planning. This study also used the Boolean operator (within each axis, we combined keywords with the Boolean operators “OR” and we then linked the search strategies for the two axes with the “AND” operator).

Study selection and eligibility criteria

This systematic review and meta-analysis focused on studies conducted in the Amhara Region, examining studies related to RTAs, mortalities and their risk factors. Studies were included based on the following criteria: (1) published in English since 2020 and (2) focused on human population RTAs and risk factors. Studies not conducted in the Amhara region, lacking relevant data on RTAs or failing to meet quality assessment criteria were excluded. Four authors independently evaluated the studies against these criteria to ensure consistency, resolving any discrepancies through discussion.

Data extraction procedure

To extract data from selected studies, a standardised Microsoft Excel format was created. This format considered different variables of interest, including first author, publication year, zone (sub-region of Amhara region), study design, sample size and key findings (online supplemental table S2). Two reviewers (HDM and KTG) independently extracted all the essential data. Any disagreement between the two authors due to inclusion and data collection was solved by the third and fourth author (ZMW and BT) through verification and discussion. All the identified potentially relevant studies were retrieved and managed using EndNote X21.

Outcome variable

The primary outcome variables for this systematic and meta-analysis were prevalence of death and injury due to RTAs in the Amhara region of Ethiopia, and the second outcome was risk factors for RTAs.

Study quality and validity

The study quality was assessed using the Joanna Briggs Institute critical appraisal checklist for studies reporting prevalence data, with high quality studies considered.¹⁶ To assess the quality of the original papers, reviewers used a blinded review approach, implementing the checklist as specified in the protocol. Articles scoring 8 or above were considered of high quality and were therefore included in the review. When there were discrepancies in the quality valuation, the third and fourth authors were consulted to resolve the issue (online supplemental table S3). This study quality assessment approach was conducted after studies met the inclusion criteria through title/abstract and full-text screening.

Data processing and analysis

The cleaned version of Microsoft Excel data was exported to STATA V.14 Software for further statistical analysis. We conducted a meta-analysis using the metan command in STATA to compute the pooled prevalence of death and injuries due to RTAs with a 95% CI. Due to the observed heterogeneity across the studies, a random-effects model was performed. Heterogeneity across studies was measured using the I² statistic, where values exceeding 50% indicated considerable heterogeneity¹⁷ and validated the choice of the random-effects approach.

We summarised the data in tables and narrative texts. Forest plots were generated for gaining insights into the overall prevalence of death and injuries due to RTAs. Subgroup analyses were conducted to explore variations in deaths and injuries due to RTAs across different zones (sub-region) of the Amhara region, enhancing our understanding of the risk factors. Additionally, publication bias was examined through funnel plots alongside statistical tests developed by Egger and Begg.¹⁸ To explore potential sources of variability, we performed meta-regression analysis using the metareg command. In this study, the outcome variable was the prevalence of death and injury due to RTAs, while the independent variables included publication year, study design, zone (sub-region) and sample size.

Patient and public involvement

As this research is a systematic review and meta-analysis of the published literature, no patients or the public were involved in the completion of this work.

Results

Description of the identified studies

This systematic review included published studies conducted on the prevalence of RTAs, injury and deaths in the Amhara region. A total of 29 records were retrieved through electronic database searching, including 25 articles and 2 grey literatures. After removing 10 duplicates, 17 articles remained for screening. A review of the titles and abstracts led to the exclusion of five articles. On full-text assessment of papers, 12 studies met the criteria for inclusion in the systematic review and meta-analysis, which were conducted from 10 January 2025 to 25 March 2025 (figure 1).

Figure 1. PRISMA flow diagram of study selection in the Amhara region, from 10 January 2025 to 25 March 2025. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Characteristics of included studies in the systematic review and meta-analysis

A comprehensive analysis was conducted on a sample of 5208 participants to determine the pooled prevalence of death and injuries due to RTAs in the Amhara region. 42% were conducted in West Gojjam, followed by East Gojjam, North Gondar and Awi special zone. The highest sample size was 1411, recorded in a study conducted in East Gojjam. Almost 99% of the studies used cross-sectional study design. The highest prevalence rate of death (58.3%) and injury (76.5%) was found in research conducted in North Gondar and West Gojjam, respectively. The poor response rate (70.5%) was reported in research conducted in Bahir Dar city (online supplemental table S2).

Pooled prevalence of death due to RTA in the Amhara region

The overall pooled prevalence of death due to RTAs in the Amhara region from random-effect model was 33% (95% CI 21% to 46%) with a heterogeneity of (I²=99.2%). This estimate was statistically significant with a p value of <0.001 (figure 2).

Figure 2. Pooled prevalence of death due to road traffic accidents in the Amhara region, Ethiopia.

Subgroup analysis of death due to RTAs in the Amhara region disparities by zone (sub-region)

Significant heterogeneity was detected among the included studies in this analysis (I²=99.06%, p < 0.01). Subgroup analysis revealed varying levels of RTA death prevalence across regions. The North Gondar showed the highest magnitude, with a rate of 55% (95% CI 51% to 59%), while Bahir Dar city showed the lowest death prevalence, recorded at 10% (95% CI 8% to 13%) (figure 3).

Figure 3. RTAs death in the Amhara region by sub-region. RTAs, road traffic accidents.

Sensitivity analyses of death due to RTAs in the Amhara region

To examine the possible influence of individual studies on the overall pooled prevalence effect, a leave-one-out sensitivity analysis was performed. The results showed that no specific study had a significant impact on the overall pooled prevalence of RTAs in the Amhara region and the combined effect remained unchanged even after excluding a specific study (online supplemental figure S1).

Publication bias of the studies on death due to RTAs in the Amhara region

The presence of publication bias was assessed using Egger’s and Begg’s regression test at a significant level of 5%. The result shows there was no statistical evidence to support the existence of publication bias, with Begg’s and Egger’s tests yielding p values of 0.24 and 0.28, respectively. Consequently, the test results provide no evidence of a small study effect (table 1). Additionally, a trim-and-fill analysis found no missing studies, further supporting the absence of publication bias. The funnel plot also performed and shows no evidence of publication bias (online supplemental figure S2).

Table 1. Publication bias assessment between RTA studies in the Amhara region, Ethiopia.

Begg’s test
adj. Kendall’s Score (P-Q)=15
SD of score=12.85
Number of studies=11
Z=1.17
Pr>|Z|=0.243
Z=1.09 (continuity corrected)
Pr>|Z|=0.276 (continuity corrected)
Egger’s test
SE	Coef.	SE	T	P>|t|	(95% CI)
Slope	0.125	0.214	0.59	0.572	−0.359 to 0.610
bias	9.338	11.397	0.82	0.434	−16.443 to 35.119

RTA, road traffic accident.

Meta-regression of studies on prevalence of death due to RTAs in the Amhara region

Meta–regression analysis was performed to investigate the relationship between factors (such as publication year, study design (cross-sectional, cohort), zone (Awi special zone, East Gojjam, West Gojjam, North Gondar, South Gondar, North Wollo, South Wollo, North Shewa, Wag Hemra, Oromia special zone and Bahir Dar special zone as Ethiopian administrative classification of Amhara region) and sample size) concerning death prevalence due to RTAs. The findings suggest that the factors do not influence the overall death prevalence (online supplemental table S4).

Pooled prevalence of RTA injuries in the Amhara region

The overall pooled prevalence of injury due to RTA in the Amhara region was estimated using a random-effects model as 48% (95% CI 36% to 60%). This estimate was statistically significant with a p value of <0.001. Higher heterogeneity was also observed across studies (I²=98.8%) (figure 4).

Figure 4. Pooled prevalence of injury due to road traffic accidents in the Amhara region.

Subgroup analysis of injury due to RTAs in the Amhara region disparities by zone (sub-region)

Significant heterogeneity was detected among the included studies in this analysis (I²=98.8%, p < 0.01). Subgroup analysis revealed varying levels of RTA injury prevalence across regions. The West Gojjam showed the highest magnitude, with a rate of 76% (95% CI 71% to 81%), while Bahir Dar city showed the lowest injury prevalence, recorded at 26% (95% CI 23% to 29%) (online supplemental figure S4).

Sensitivity analyses of injury due to RTAs in the Amhara region

To examine the possible influence of individual studies on the overall pooled prevalence effect, a leave-one-out sensitivity analysis was performed. The results showed that no specific study had a significant impact on the overall pooled prevalence of RTAs in the Amhara region and the combined effect remained unchanged even after excluding a specific study (online supplemental table S5).

Publication bias of the studies on RTAs and injury in the Amhara region

The presence of publication bias was assessed using Egger’s and Begg’s regression test at a significant level of 5%. The result shows there was no statistical evidence to support the existence of publication bias, with Begg’s and Egger’s tests yielding p values of 0.48 and 0.53, respectively. Consequently, the test results provide no evidence of a small study effect (online supplemental table S6). The funnel plot also suggests there is no evidence of publication bias and it is symmetric (figure 5).

Figure 5. Funnel plot for publication bias assessment of injury due to RTA studies in the Amhara region. RTA, road traffic accident.

Factors associated with RTA in the Amhara region

Sociodemographic-related driver characteristics

The drivers’ age significantly impacted RTAs. Young drivers were associated with a higher likelihood of RTAs.^{9 11 12 19 20} Drivers’ experience has a significant effect on the likelihood of RTAs.²⁰ Experienced drivers are expected to have better knowledge about the road condition, safety and traffic signs, and rules tend to cause fewer traffic accidents. Drivers with little experience cause more accidents. On the other hand, drivers’ education level is also a factor for RTAs,^{20 21} as drivers’ education level of grade 12 and above the likelihood of RTAs is low. Sex of drivers is also a cause for RTAs.¹¹

Driving behaviour related to drivers’ characteristics

Drivers’ behaviour was the major responsible cause for death and injury due to RTAs in Amhara region. These behavioural factors included over-speeding, phone use, tailgating, not indicating, dangerous overtaking, overloading, driving under alcohol, driving under fatigue and failure to yield to pedestrians.^{4 9 11 12 19 21–24} Moreover, non-adherence to road safety protocols was a crucial determinant of accident severity. These behaviours increased the likelihood of crash injury severity and were associated with a higher magnitude of RTA involvement.

Vehicle-related factors

The vehicle factor is also another responsible factor for RTAs in the Amhara region. The longer the service time the vehicle provides, the higher the chance of the occurrence of mechanical defects, which in turn cause RTAs.^{9 20} Ownership of the vehicle is also a contributing factor to RTAs. Employed drivers were associated with a higher likelihood of RTAs.²⁰ Hence, drivers are not filling the sense of ownership of the properties. Mostly public vehicles (Isuzu, Isuzu Track, Abdulla) were a responsible factor for death due to RTAs.^{9 21} Licence grades are also another important factor for RTAs.²¹

Environmental and road conditions related factors

Environmental factors like season, weekend day, time of day and light conditions also have a great contribution to RTAs. One might assume that accidents happen more in the wet seasons.^{20 22} Generally, the incidence of accidents is surprisingly high during the day due to heavy traffic movement (vehicles and pedestrians), then the night. Road conditions are also the major contributing factor for RTAs in the Amhara region.^{9 12} Most RTAs occurred in street roads in the Amhara region in West, East Gojjam and Awi special zones, given that the quality of the street was not considered in this study.

Discussion

This systematic review and meta-analysis aimed to investigate the pooled prevalence of RTA and their associated factors in the Amhara region. Taking stock of existing systematic reviews and meta-analyses is an important approach to use in indicating gaps in new research and to enhance evidence-based decision-making. A total of 12 studies were included in the review. The overall pooled prevalence of death and injury due to RTAs in the Amhara region was found to be 33% and 48%, respectively. In this systematic and meta-analysis, driver’s age, education, experience, ownership, season, vehicle services age, daylight condition, road condition and driver-related behavioural factors, including overloading, speeding, failure to give priority to pedestrians and driving under the influence of alcohol, were the identified risk factors for RTAs.

The subgroup meta-analysis result indicates that the highest pooled death estimates of RTAs were seen in North Gondar and west, East Gojjam and Awi 58%⁹ and the lowest pooled estimate 7%²³ in Bahir Dar Town. The Meta-regression analysis indicated that sample size, publication year, study design and Zone (sub-region) had no significant impact on RTA. These results might be masking due to small study effects. Research shows that small sample size is a cause for hiding the true variations.^{25 26}

In this study, the overall magnitude of death and injury due to RTA in the region is more severe than Tigray (23.2%),²⁷ Somali²⁸ and Afar.²⁹ It is also higher than the average national level RTA induced mortality (27.7%).²¹ It is also higher than in South Sudan.³⁰ The discrepancy might be due to population size proportionality and road infrastructure variations, like traffic signals at intersections, which help ensure driver compliance with regulations and reduce accident likelihood. Moreover, the interplay between socioeconomic conditions and infrastructure development plays a critical role in shaping the quality of roads, ultimately impacting overall road safety and efficiency.

Drivers’ age significantly affects the occurrence of RTAs, with younger and older drivers showing higher risk patterns.^{9 11 12 19 20} These studies are consistent with other studies conducted.^31–33 A possible reason might be that younger drivers lack sufficient driving exposure, which leads to poor hazard perception, higher likelihood of speeding, tailgating or illegal overtaking, drug impairment and underdeveloped prefrontal cortex, which reduces impulse control. To overcome these issues, research outputs recommend that graduated licensing,^{34 35} telematics monitoring^{36 37} and mandatory hazard training will reduce these risks.

Drivers’ education level is also a significant risk factor for the occurrence of RTAs. Research consistently shows that drivers with lower or no formal education have a higher risk of being involved in fatal RTAs compared with those with higher educational attainment.^{31 38 39} Trained drivers had more behavioural changes to obey traffic rules and make safer decisions on the road. In addition, education enhances their knowledge and awareness of road safety and provides access to better resources, such as safer vehicles, further contributing to a lower risk of accidents.⁴⁰ Though there is no enough evidence on drivers’ sex with relation to RTAs, female drivers have tended to cause low traffic accidents.¹¹

Driver’s behaviour was the major responsible cause for death and injury due to RTAs in the Amhara region. These behavioural factors included over-speeding, phone use, tailgating, not indicating, dangerous overtaking, overloading, driving under alcohol, driving under fatigue and failure to yield to pedestrians.^{4 9 11 12 19 21–24} Moreover, non-adherence to road safety protocols was a crucial determinant of accident severity. These behaviours increased the likelihood of crash injury severity and were associated with a higher magnitude of RTA involvement.⁴¹

The occurrence of RTAs can be significantly influenced by the time of day due to various factors such as visibility, traffic volume, driver behaviour and fatigue.⁴² Seasonal (weather conditions) have a significant impact on both the occurrence and severity of RTAs. In this study, accidents happen more in the wet seasons.^{20 22} Generally, the incidence of accidents is surprisingly high during the day due to heavy traffic movement (vehicles and pedestrians) than during the night.^{9 20} However, studies in the USA highlight that RTAs are higher in snow/icy than dry conditions.⁴³ Different seasons bring about varying environmental conditions that can affect road safety in different ways.^{44 45} Road type and condition are also a responsible factor for RTAs.^{14 42 44–46} The overall prevalence of death and injuries due to RTA is influenced by various social and economic factors in the region and, in turn, these accidents impose substantial social and economic burdens. Thus, the region needs to intervene by designing policies and implementing strategies.

Conclusion

This study mainly focused on the pooled prevalence and risk factors of RTA injuries and mortalities in the Amhara region, Ethiopia. The research emphasises the need for governments to proactively address RTAs through comprehensive strategy, focusing on strict enforcement of traffic laws, especially against common violations like speeding, younger age and ignoring signals. Governments ought to invest in road design, in extensive driver training programmes to improve awareness of road safety and promote responsible driving behaviours. Besides, the government should monitor the implementation of traffic rules in all administrative levels and strengthen the integration among concerned NGOs, civic societies and the community and drivers closely working with road safety security. Implementing these strategies will have the potential to decrease the incidence of RTAs.

Limitation of the study

This research represents a pioneering systematic review and meta-analysis aimed at estimating the aggregated prevalence and risk factors of RTAs across the Amhara region. Emphasis on English-language publications in the study introduces a potential language bias, thereby constraining the range of research examined. Although Egger’s test was employed to identify publication bias, its validity may be undermined by the small number of studies analysed, as those yielding positive results are often more likely to be published. The existence of unexamined heterogeneity, despite the execution of subgroup analyses, indicates that there may be other factors affecting the outcomes that have not been considered. Since the number of studies is small, subgroup analysis results may not be reliable. Variables measurement standardisation issue masks the finding of the pooled estimate between studies.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.