Epidemiology of fatal motorcycle crashes in Ghana: a 23-year analysis of national crash data

Abstract

Background

Motorcycles are a common mode of transport in Ghana, such that in some regions, they account for 90% of all registered vehicles. Their popularity and associated crash vulnerabilities have become a significant public health issue. However, there is limited knowledge of the long-term epidemiological profile of fatal crashes in Ghana. This study examines the epidemiology of fatal motorcycle crashes over a 23-year period.

Methods

We analysed 23 years of national data on motorcycle crashes. The data is collated by the Building and Road Research Institute from police-reported crash investigation files. Data were summarised for the temporal, environmental, and rider-related features of fatal crashes. A mixed-effects Poisson regression with robust variance estimation was employed to identify factors influencing fatal motorcycle crashes.

Results

Over the 23 years, 40,322 motorcycle crashes occurred, and 22.85% were fatal. Fatal crashes have increased, with an average annual percent change increase of 4.71% and 60% occurring from 2016 to 2022. The average age of the riders was 31.64 years, and 98.90% of the fatal crash riders were males. A greater proportion of fatality ratios were recorded in the northern parts Ghana (Upper West = 36.80% and Upper East = 36.60%), and during the last three months of the year (28.97%). The day of the week, time of day, weather, traffic control, road type, rider age and sex, and collision type were significant factors associated with fatal motorcycle crashes.

Conclusion

The findings emphasise a significant public health burden from motorcycles, particularly over the past 10 years. Targeted interventions (e.g., speed-control measures) are needed to address this crisis, particularly as the country prepares to commercialise its use for passenger transport.

Introduction

Globally, fatalities involving motorcycles account for 21% of all road traffic fatalities, with a significant proportion occurring in low- and middle-income countries (LMICs) [1]. While all World Health Organisation (WHO) regions observed a decrease in estimated road traffic fatalities during the UN first decade of action for road safety initiatives, the African region experienced a significant increase of 17% [1]. During the same period, fatalities involving motorcycles in the region doubled [2]. The reliance on these vehicles has become essential, and in some communities, they are often the only means of transportation [3]. Their growing popularity as an alternative means of transport can be attributed to several factors, including poor public transport services, affordability, ease of maintenance, convenience, and the ability to manoeuvre during peak hours [4–6].

In Ghana, motorcycles have become one of the most prevalent modes of transport. Since 2010, it has provided employment for many through its commercialisation for passenger transport and the delivery of goods, although the law prohibited its use for passenger transport until December 2025. The increase in motorcycle usage has also come with a significant rise in road traffic crashes and fatalities. Evidence shows that motorcyclists are three times and sixteen times more likely to be injured and die, respectively, during a crash than vehicle occupants [7]. The substantial increase in motorcycle crashes and fatalities in Ghana has become a major public health issue, highlighting the urgent need for systematic, long-term crash data analysis to guide effective and targeted interventions.

Although studies have reported on the severity of motorcycle injuries in the country, to the best of our knowledge, no comprehensive longitudinal analysis of fatal motorcycle crashes has been conducted in the country. Our study addresses this gap by analysing 23 years of national crash data to examine the epidemiology of fatal motorcycle crashes in Ghana. We assess the prevalence and explore the key contributing factors. The insights from this analysis will help inform and tailor interventions and policies to curb the rising toll of motorcycle fatalities in Ghana.

Methods

Data source

The data used for this study were obtained from the national crash database maintained by the Council for Scientific and Industrial Research (CSIR) Building and Road Research Institute (BRRI). This database collates information from police files, compiled by trained technicians specialising in road traffic safety, using a standardised case report form. The case report form includes details on the nature of the crash, its geographic location, the vehicles involved, and casualties, including passengers, pedestrians, and motorcyclists. The form is designed to meet the standards required by the Africa Transport Policy Program (SATPP) for crash investigation [8]. The gathered data is coded and stored using the Micro-computer Accident Analysis Package (MAAP) software, developed by the Transport Research Laboratory (TRL) in the United Kingdom. The motorcycle crash data retrieved for this study covers the period from 2000 to 2022.

Study variables

Outcome variable

The outcome variable for this study was fatal motorcycle crashes. This was measured as a binary variable, with “0” indicating a non-fatal motorcycle crash (comprising severe injury, minor injury, and damage-only crashes) and “1” indicating a fatal motorcycle crash.

Predictor variables

The exposure (explanatory) variables in this study were crash characteristics (for example, collision type), riders’ characteristics (such as age, sex, and riders’ error), road characteristics (road width, location, and road description, etc.), time characteristics (day of week, time of day, months, etc.).

Statistical analysis

Descriptive statistics were employed to summarise the data. Means and standard deviations were used to summarise numerical variables, whereas frequencies and percentages were applied to summarise categorical variables. The regional fatality ratio for motorcycle crashes was estimated by dividing the number of fatal crashes in each region by the total number of reported crashes in that region. Jointpoint regression was used to identify a significant trend in fatal motorcycle crashes over the 23 years.

Univariate and multivariable mixed-effect Poisson regression were employed to identify factors associated with fatal motorcycle crashes, accounting for both between- and within-region and year variability. Mixed-effect Poisson regression was used to estimate the risk ratio (RR). Despite the binary outcome, this approach is recommended for estimating RR [9]. In the mixed-effects Poisson regression, year and region were specified as random effects to account for unobserved heterogeneity (variability) across time and geography. This approach allowed us to adjust for clustering while avoiding over-parameterisation that could arise from modelling year as a fixed effect over the extended study period. At the univariate level, variables with p-values of 0.2 (80% α-level) were included in the multivariable analysis. Additionally, variables deemed potential confounders but not statistically significant in the univariate analysis were adjusted for in the multivariable analysis. An automated stepwise approach was utilised to incorporate variables into the multivariable mixed-effect Poisson regression model. Variables were re- entered and rechecked until a parsimonious model was achieved. Akaike’s Information Criterion (AIC) and likelihood-ratio tests were employed to compare nested candidate models and inform the selection of the best-fitting model. A 95% confidence interval was computed, with p-values < 0.05 considered statistically significant. Data were analysed using Stata version 18.0, and the results were presented in tables and charts.

Results

Table 1 presents the crash and rider characteristics of fatal motorcycle crashes in Ghana from 2000 to 2022. Over the 23 years, 40,322 motorcycle crashes occurred in Ghana, of which 9,213 (22.85%; 95% CI: 22.44% to 23.26%) were fatal. A higher proportion of the fatal crashes were head-on (35.10%), rear-end (15.83%), and pedestrian-hit (15.42%). Hit-and-run fatal crashes were 3.23%. Four-fifths of fatal motorcycle crashes occurred in areas where there was no traffic control (80.35%).

Table 1.

Crash and rider characteristics of fatal motorcycle crashes, 2000 to 2022

Variables	Crash severity (N = 40,322)
	Fatal (n = 9,213)		Non-fatal (n = 31,109)
	Frequency	Percentage	Frequency	Percentage
Crash characteristics
Collision type
Head on	3,233	35.09	4,690	15.08
Rear end	1,458	15.83	7,615	24.48
Right Angle	680	7.38	4,149	13.34
Side Swipe	869	9.43	6,215	19.98
Ran off-road	932	10.12	1,256	4.04
Hit an object on the road	50	0.54	88	0.28
Hit an object off-road	129	1.40	134	0.43
Hit parked vehicle	254	2.76	668	2.15
Hit Pedestrian	1,421	15.42	5,869	18.87
Hit Animal	72	0.78	244	0.78
Other	115	1.25	181	0.58
Hit and run
No	8,915	96.77	29,759	95.66
Yes	298	3.23	1,350	4.34
Traffic control
Yes	554	6.01	4,074	13.10
No	7,403	80.35	22,016	70.77
Not recorded	1,256	13.63	5,019	16.13
Rider’s characteristics
Age, mean (SD)	31.64 (10.47)		30.8 (9.35)
Age group
< 18 years	224	2.43	674	2.17
18–29 years	4,168	45.24	14,768	47.47
30–39 years	3,077	33.40	10,619	34.13
40–49 years	1,014	11.01	3,479	11.18
50 and above years	730	7.92	1,569	5.04
Sex
Male	9113	98.91	30,717	98.74
Female	100	1.09	392	1.26
Contributing factor
None	2,113	22.93	9,004	28.94
Inexperience	219	2.38	924	2.97
Inattentive	3,090	33.54	12,160	39.09
Too Fast	2,567	27.86	5,215	16.76
Too Close	117	1.27	686	2.21
No Signals	36	0.39	152	0.49
Improper Overtaking	388	4.21	810	2.60
Improper Turning	100	1.09	389	1.25
Other	583	6.33	1,769	5.69

SD= Standard deviation. Note, the contributing factors were reported by the police based on their crash investigation

The average age of the riders was 31.64 years, with a standard deviation of 10.47 years. A higher proportion (73.15%) of riders were under 36 years of age, and a majority (98.91%) were male. Riders’ inattention and speeding accounted for 61.40% of the fatal crashes.

On the nature of roads where fatal crashes occurred, 86.56% were straight and flat roads, 30.40% were roads with potholes, and 99.26% of the road surfaces were dry. The average road width was 7.59 m (standard deviation = 1.31 m). Approximately 15% of the crashes occurred at junctions, including 4-way junctions, T-junctions, and Y-junctions (Table 2).

Table 2.

Road (place) characteristics of fatal motorcycle crashes, 2000 to 2022

Variables	Crash severity (N = 40,322)
	Fatal (n = 9,213)		Non-fatal (n = 31,109)
	Frequency	Percentage	Frequency	Percentage
Road description
Straight and flat	7,975	86.56	28,816	92.63
Curve only	689	7.48	1,183	3.80
Incline only	146	1.58	426	1.37
Curve and incline	363	3.94	630	2.03
Bridge	40	0.43	54	0.17
Road surface
Tar, good	6,207	67.37	23,677	76.11
Tar, potholes	2,227	24.17	5,979	19.22
Gravel	205	2.23	417	1.34
Earth, potholes	574	6.23	1,036	3.33
Road width, means (SD)	7.59 (1.31)		7.72 (1.28)
Road surface condition
Dry	9,145	99.24	30,930	99.42
Wet	55	0.62	161	0.52
Muddy	13	0.14	18	0.06
Type of location
Not at a junction	7,757	84.20	22,549	72.48
At a junction	1,399	15.19	8,265	26.57
Not recorded	57	0.62	295	0.95
Type of community
City	4,038	43.83	22,021	70.79
Town	446	4.84	849	2.73
Village	4,729	51.33	8,239	26.48

SD= Standard deviation

For regions, 16.21% of the fatal crashes were from Ashanti (16.21%), Greater Accra (14.64%), Brong Ahafo (now divided into three regions: Bono, Bono East, and Ahafo) (12.67%), and Western (10.34%) regions (data not shown). The fatality ratio for crashes was higher in the Upper West (36.8%), Upper East (36.6%), and Brong Ahafo (33.5%) (Fig. 1).

Fig. 1.

Fatality ratio from motorcycle crashes in Ghana by regions, 2000 to 2022

Over 60% of the fatal motorcycle crashes occurred from 2016 to 2022. Fatal motorcycle crashes have increased over the past 23 years (Fig. 2), with an average annual percent change (AAPC) of 4.71% (95% CI: 3.00 to 6.45). A higher proportion (28.97%) of fatal crashes occurred during the last three months of the year (October to December). A greater percentage (34.5%) of the fatal crashes occurred in the evening (18:00–21:59). The light condition with the highest proportion of fatal crashes was during the day (59%), and the weather condition with the highest proportion of fatal crashes was clear weather (85.7%) (Table 3).

Fig. 2.

Joinpoint regression of fatal crashes per 1,000 motorcycle crashes, 2000–2022

Table 3.

Times of fatal motorcycle crashes in Ghana, from 2000 to 2022

Variables	Crash severity (N = 40,322)
	Fatal (n = 9,213)		Non-fatal (n = 31,109)
	Frequency	Percentage	Frequency	Percentage
Year
2000–2005	484	5.25	3,260	10.48
2006–2010	1,132	12.29	4,694	15.09
2011–2015	2,040	22.14	6,357	20.43
2016–2020	3,539	38.41	10,942	35.17
2021–2022	2,018	21.90	5,856	18.82
Months
January – March	2,054	22.29	6,793	21.84
April – June	2,256	24.49	7,680	24.69
July – September	2,234	24.25	7,824	25.15
October – December	2,669	28.97	8,812	28.33
Day of the week
Weekdays	6,282	68.19	22,525	72.41
Weekends	2,931	31.81	8,584	27.59
Time of day
Morning (6:00–11:59am)	2,015	21.87	9,013	28.97
Afternoon (12:00–17:59pm)	3,108	33.73	12,677	40.75
Evening (18:00–21:59pm)	3,179	34.51	7,571	24.34
Night (22:00pm-5:59am)	911	9.89	1,848	5.94
Weather
Clear	7,897	85.72	28,674	92.17
Fog/mist/dust/smoke	327	3.55	611	1.96
Rain	37	0.40	106	0.34
Dazzle	166	1.80	176	0.57
Others	786	8.53	1,542	4.96
Light condition
Day	5,444	59.09	22,470	72.23
Nights with no light	1,642	17.82	2,853	9.17
Nights with light on	2,009	21.81	5,311	17.07
Not recorded	118	1.28	475	1.53

Factors influencing fatal motorcycle crashes in Ghana

Table 4 presents a mixed-effects Poisson regression showing the factors associated with fatal motorcycle crashes. In the bivariate analysis, factors such as the day of the week, time of day, weather, light condition, traffic control, nature of the road, location of the crash, age of the crash victim, sex, error committed by the rider, and collision type were significantly associated with fatal motorcycle crashes.

Table 4.

Mixed-effect Poisson regression of factors influencing fatal motorcycle crashes in Ghana, 2000–2022

Variables	cRR	95% CI	aRR	95% CI
Day of week
Weekday	1	1	1	1
Weekend	1.14	1.085, 1.202	1.08*	1.044, 1.123
Time of day
Morning	1	1	1	1
Afternoon	1.04	1.010, 1.088	1.04	0.994, 1.078
Evening	1.47	1.374, 1.577	1.29*	1.192, 1.387
Night	1.80	1.579, 2.047	1.50*	1.347, 1.672
Weather
Clear	1	1	1	1
Fog/mist/dust/smoke	1.55	1.355, 1.774	1.09	0.981, 1.216
Rain	1.15	0.826, 1.599	1.06	0.806, 1.394
Dazzle	1.79	1.548, 2.074	1.13*	1.004, 1.271
Others	1.47	1.362, 1.590	1.06	0.987, 1.141
Light condition
Day	1	1	1	1
Nights with no light	1.68	1.582, 1.785	1.16*	1.079, 1.245
Nights with light on	1.37	1.209, 1.548	0.95	0.868, 1.037
Traffic control
No	1	1	1	1
Yes	0.63	0.548, 0.736	0.82*	0.737, 0.909
Road description
Straight and flat	1	1	1	1
Curve only	1.50	1.331, 1.683	1.27*	1.187, 1.352
Incline only	1.08	0.905, 1.286	1.00	0.857, 1.158
Curve and Incline	1.50	1.322, 1.699	1.24*	1.109, 1.393
Bridge	1.64	1.238, 2.176	1.27*	1.006, 1.612
Location type
Not at a junction	1	1	1	1
At a junction	0.67	0.630, 0.705	0.81*	0.786, 0.846
Age group
< 18 years	1.05	0.919, 1.203	1.06	0.956, 1.182
18–29 years	1	1	1	1
30–39 years	1.05	1.020, 1.091	1.07*	1.026, 1.096
40–49 years	1.03	0.940, 1.096	1.03	0.967, 1.102
50 and above years	1.27	1.181, 1.466	1.39*	1.258, 1.536
Sex
Male	1	1	1	1
Female	0.68	0.606, 0.757	0.83*	0.728, 0.936
Contributing factor
None
Inexperience	1.02	0.915, 1.128	0.96	0.876, 1.047
Inattentive	1.03	0.969, 1.099	0.99	0.945, 1.035
Too Fast	1.67	1.520, 1.832	1.45*	1.297, 1.623
Too Close	0.79	0.693, 0.891	0.86*	0.752, 0.993
No Signals	0.91	0.604, 1.377	0.99	0.633, 1.538
Improper Overtaking	1.64	1.521, 1.759	1.49*	1.416, 1.577
Improper Turning	1.03	0.858, 1.171	1.14	0.971, 1.332
Other	1.32	1.188, 1.457	1.10*	1.022, 1.178
Collision type
Head on
Rear end	0.44	0.388, 0.492	0.49*	0.434, 0.547
Right Angle	0.41	0.362, 0.460	0.51*	0.468, 0.575
Side Swipe	0.33	0.300, 0.372	0.37*	0.330, 0.410
Ran off-road	1.03	0.940, 1.130	0.95	0.870, 1.040
Hit an object on the road	0.89	0.638, 1.247	0.80	0.578, 1.097
Hit an object off-road	1.30	1.077, 1.560	1.19	0.990, 1.437
Hit parked vehicle	0.69	0.590, 0.812	0.68*	0.598, 0.780
Hit Pedestrian	0.55	0.488, 0.621	0.58*	0.509, 0.652
Hit Animal	0.55	0.453, 0.676	0.63*	0.518, 0.761
Other	0.97	0.769, 1.230	0.97	0.772, 1.231

Categories of some variables labelled “Not recorded” were treated as distinct categories in the analysis but are not presented in the regression tables for clarity

cRR= crude Risk Ratio, aRR= adjusted Risk Ratio, CI= Confidence Interval, *=significant at 5% alpha level

The multivariable analysis revealed that the risk of fatal motorcycle crashes was 8% higher on weekends than on weekdays (aRR = 1.08, 95%CI = 1.04 to 1.12). Fatal motorcycle crashes were 29% and 50% higher in the evening and at night compared with the morning (aRR = 1.29, 95%CI = 1.19 to 1.39; aRR = 1.50, 95%CI = 1.35 to 1.67), respectively. Fatal motorcycle crashes were 16% higher during nights with no lights compared with during the day (aRR = 1.16, 95%CI = 1.08 to 1.25), and 18% less when there was traffic control (aIRR = 0.82, 95%CI = 0.74 to 0.91).

The fatality risk of motorcycle crashes was 27% higher on both curved-only sections of roads and bridges (aRR = 1.27, 95%CI = 1.19 to 1.35; aRR = 1.27, 95%CI = 1.01 to 1.50, respectively), and 24% higher on combined curved and inclined sections of roads (aRR = 1.24, 95%CI = 1.11 to 1.39) compared with straight and flat roads. The fatality risk was 19% lower when the crash happened at a junction compared with non-junction locations (aRR = 0.81, %CI = 0.79 to 0.85). The fatality risk of motorcycle crashes was 17% lower among female riders compared with males (aRR = 0.83, %CI = 0.73 to 0.94), 7% higher in motorcyclists who were aged 30–39 years (aRR = 1.07, %CI = 1.03 to 1.10), and 39% higher in those above 50 years (aRR = 1.39, %CI = 1.26 to 1.54) compared with crashes among riders aged 18–29 years of age. The risk of fatal motorcycle crashes were 45% more among riders riding too fast (aRR = 1.45, %CI = 1.30 to 1.62), 49% higher when engaged in improper overtaking, but 14% lower when riding too close to each other (aRR = 0.86, %CI = 0.75 to 0.99) compared with riders who committed no error.

Fatal motorcycle crash risk was significantly lower in all types of collisions than head-on collisions: 51% lower for rear-end collisions (aRR = 0.49, %CI = 0.43 to 0.55), 49% lower for right-angle collisions (aRR = 0.51, %CI = 0.47 to 0.58), 63% lower for side-swipe collisions (aRR = 0.37, %CI = 0.33 to 0.41), 32% lower when the motorcycle hit a parked vehicle (aRR = 0.68, %CI = 0.60 to 0.78), 42% lower when a pedestrian was hit (aRR = 0.58, %CI = 0.51 to 0.65), and 37% lower when the collision involved an animal (aRR = 0.63, %CI = 0.52 to 0.76).

Discussion

The study investigated the long-term epidemiological profile of fatal motorcycle crashes in Ghana, which, to our knowledge, remains underexplored. We employed a mixed-effects analysis to account for regional and time variability, and identify factors that influence fatal motorcycle crashes nationwide. Our findings revealed that 22.85% of all motorcycle crashes over the 23-year period were fatal, with 60% of these crashes occurring between 2016 and 2022. Geographically, the highest share of motorcycle fatalities occurred in the northern parts of the country. Fatal motorcycle crashes were also higher during the last three months of the year and at weekends. Factors such as day of the week, time of day, weather, lighting, traffic control, road type and description, age and sex of rider, and collision type were significantly associated with fatal motorcycle crashes.

The estimated prevalence of fatal motorcycle crashes in this study is high. At least 23 out of every 100 motorcycle crashes resulted in fatalities over the past 23 years. This high prevalence is concerning but not surprising, as helmet use in the country has been reported to be low [4, 10, 11]. Research shows that helmet use reduces fatalities by 28% to 64% [12], suggesting that the country’s low helmet use may have contributed to this high prevalence. Although we did not measure helmet use in this study, utilisation in Ghana remains low [5, 11], despite helmets being mandatory for motorcyclists under regulation. Interventions to increase helmet use are necessary to help reduce fatal motorcycle crashes. Fatal motorcycle crashes have seen a substantial increase in Ghana over the past two decades, with a peak increase from 2016. Over 60% of fatal crashes were recorded between 2016 and 2022. While our findings show an increase in the proportion of fatal motorcycle crashes over the study period, particularly between 2016 and 2022, these should be interpreted with caution. The observed rise may reflect not only a true increase in fatal crashes but also shifts in underlying exposure, such as the substantial increase in motorcycle use over the past decade [3], as well as other factors. Ideally, temporal trends should be expressed as rates (e.g., fatalities per 10,000 registered motorcycles per region-year) to account for changes in exposure. However, the absence of comprehensive data on motorcycle registration limited our ability to compute exposure-adjusted rates. Although we recognise the limitations of relying solely on counts, the available data still provide essential insights into the growing burden of motorcycles. In major urban areas, motorcycles are used for deliveries, and in rural areas for passenger transport. In other communities, motorcycles are the only form of public transportation available. This increase in its use, along with riders’ disregard for road safety regulations, may have contributed to the rise in fatal motorcycle crashes during this period.

Riders’ age and sex were significantly associated with fatal crashes. Riders aged 50 and above exhibited a higher probability of fatal crashes than those younger than 30 years. Although younger riders are involved in a greater number of crashes, older riders tend to experience more fatal outcomes. This finding supports the idea that older people are more vulnerable and less able to withstand severe trauma [13]. The risk of severe injury is strongly linked to age, with older people facing a greater chance of fatalities [14–16]. As a person ages, they become more physiologically vulnerable and fragile and are likely to suffer more serious outcomes from the impact of a crash. And in the event of a motorcycle crash, the outcome is more likely to be fatal. And consistent with global studies [16–19], fatal crashes were more prevalent among male riders than female riders. This disparity can be attributed to males’ greater exposure to road risks, including more frequent riding, occupational and environmental exposures, speeding, and weaker traffic enforcement [20]. Additionally, behavioural exposures include neglecting traffic regulations, engaging in aggressive driving, and often ignoring safety precautions, such as wearing a helmet.

Speeding is a well-established factor significantly associated with a higher risk of severe to fatal road crashes. Higher speeds reduce a rider’s reaction time to respond to hazards and increase the energy released in a collision. In LMICs like Ghana, speeding is frequently cited in crash analyses as a major factor in fatal and severe crashes [21]. Global evidence indicates that a small increase in speed can result in disproportionate outcomes: a 5% increase in average speed is associated with approximately a 20% increase in fatal crashes [21, 22]. This correlation is grounded in both physical and human factors. As speed increases, kinetic energy rises exponentially, resulting in greater impact forces and diminishing the rider’s ability to perceive and respond to risks [22]. Consequently, a motorcyclist travelling at high speed not only faces a heightened risk of crashing but also a significantly increased probability that the crash will be fatal. Similarly, improper overtaking has been recognised as a contributor to fatal crashes in Africa [21], reflecting the persistence of this high-risk behaviour across many African settings. The speeds of overtaking vehicles generate extreme impact forces that exceed human tolerance, leading to fatal outcomes.

Weekend and night riding were significantly associated with fatal motorcycle crashes in Ghana. Consistent with our findings, studies from the Australian State of Victoria and Thailand have reported high motorcycle fatalities on weekends [23, 24]. The high risk of fatal motorcycle crashes during weekends in Ghana could be attributed to the purpose for which people ride, such as for recreation, leisure, and funerals. Therefore, riders may engage in risky behaviours like high alcohol consumption, speeding, and non-compliance with safety measures, which increases the risk of a fatal crash [23, 25]. Similarly, evening and night riding are frequently associated with high alcohol (not measured in this study) intake and increase the risk of fatal crashes. Fell and colleagues’ observation on blood alcohol concentration (BAC) measures in fatal crashes shows that BAC was three times higher at night than during the day [26]. Additionally, at night visibility is reduced, and long-distance riders may experience fatigue, which may have contributed to the high fatality rates.

Seasonally, crashes and related fatalities peak during the last quarter of the year. Without exception, fatal motorcycle crashes are higher during this period. In Ghana, this time is characterised by festivities, holidays and travels, exposing people to risky, fatal crash behaviour, such as drinking and speeding. Also, crashes during adverse weather conditions, such as rain, fog/mist, and dazzle, were more likely to be fatal than those occurring in clear weather. The impact of weather on crash severity has produced mixed results, with some studies supporting our findings [27, 28] and others contradicting them [29–31]. Unlike motor vehicles, motorcycles offer limited protection in adverse weather, making them more vulnerable to severe crashes.

Geographically, fatal motorcycle crashes were higher in northern Ghana. In this part of the country, the traffic composition has shifted significantly since the early 2000s, with motorcycle registrations accounting for approximately 90% of all registered vehicles in 2019 [3]. Motorcycles serve as the primary mode of transportation to and from the Central Business Districts, farms, schools, and workplaces, with approximately 80% of residents owning at least one motorcycle [32]. Despite the widespread ownership and registration of motorcycles, the emphasis on motorcycle safety, such as helmet use, remains very low [5, 10, 11, 33, 34]. Northern Ghana has the lowest helmet use in the country, as reported in multiple studies [4, 5, 10, 11, 33–36]. These factors could explain why, geographically, the highest burden of fatal motorcycle crashes occurs in northern Ghana. Similar to our study, Kudebong and colleagues have found that motorcycle crashes are the leading cause of outpatient visits and hospital admissions in this same geographical area [37].

Crashes on curved road segments and bridges were associated with a higher risk of fatality than those on straight and flat roads. These roadway features likely increase crash severity by increasing the likelihood of vehicle instability or loss of control, particularly at higher speeds. Navigating a curve requires motorcyclists to lean their bikes and maintain precise control over both speed and trajectory. Any misjudgement, such as entering a turn at excessive speed or encountering a slippery surface, can result in loss of traction or balance, often leading to a slide or rollover. Such crashes are frequently severe due to the abrupt loss of control and lack of protective structures for riders. Empirical evidence shows that horizontal road curvature significantly increases the risk of severe or fatal crashes in motorcycle crashes, particularly on high-speed roads [38]. Additionally, crashes at locations without traffic control were more likely to be fatal. Suggesting that where traffic flows are unregulated, crashes tend to occur at higher speeds or in more dangerous conflict scenarios. Consistent with our study, Chandra and Mohan in India, another developing country, reported that road sections without traffic controls accounted for 79% of road traffic fatalities [39]. The absence of traffic signals or signage contributes to fatal crashes, as uncontrolled road sections can lead to high-speed conflicts, thereby increasing the risk of fatalities for riders.

On the contrary, we found that motorcycle crashes at junctions had a lower fatality risk than those at non-junction road segments. Although junctions may have more conflicts [40], they can also enforce lower approach speeds or incorporate features such as road signs, which can reduce the severity of a crash if one occurs. Although most crashes occurred on straight and flat roads, the risk of fatal crashes was higher on curved and sloped road segments. Evidence shows that curves pose a greater risk of crash severity due to loss of control and higher impact forces [41–43]. The high risk of fatal crashes on curved roads may be attributed to inadequate road signage when approaching these segments on most roads in Ghana. Although this has not been scientifically documented, radio and television stations have reported the absence of road signs at crash-prone sites along certain road sections in the country.

Our findings showed that head-on collisions were the deadliest type of crash for motorcyclists. This is intuitive and aligns with injury biomechanics: a head-on crash, especially when speeding, releases an enormous force to an unprotected rider, often resulting in severe consequences. It is not surprising that our study found speeding and improper overtaking to be significant rider errors that lead to fatal crashes, as they are also common precipitating factors in head-on collisions. Adanu and colleagues have posited that the primary phenomenon contributing to the large number of fatal head-on collisions is improper overtaking [40]. Consistent with our study’s findings, prior studies in Ghana have identified overtaking as a major contributor to fatal crashes [44–46]. A large proportion of [40] Ghana’s roads are undivided [39], which may explain instances of wrongful overtaking and the high number of fatal head-on crashes. Reengineering roads using the safer roads concept can make them more forgiving of driving errors. Additionally, earlier studies in the country recommended installing speed humps at crash-prone black spots [47] and rumble strips on road sections prone to dangerous overtaking [48, 49].

In line with the Second Decade of Action for Road Safety (2021–2030), which aims to halve global road traffic deaths and injuries by 2030, this study provides strong evidence to support targeted national interventions in Ghana. The high rate of fatal motorcycle crashes and the factors identified in Ghana emphasise the need to.◦

• ◦ strengthen speed management and traffic enforcement,

• ◦ improve traffic control, structured training programmes on defensive riding, and.

• ◦ promote the adoption of Anti-lock Braking Systems (ABS) on motorcycles.

Limitations

Some limitations should be considered when interpreting the findings of this study. The first is the underreporting of crash and fatality data in Ghana. While efforts have been made to account for under-recording, this has not been consistent, and the extent of non-reporting remains unknown, potentially leading to an underestimation of the true number of fatal crashes. The second limitation is that key variables such as helmet use, actual speed, alcohol and drug use, and kilometres travelled were not controlled for in the analysis. Although the variables included do represent significant risk factors for policy and practice, we recognise that the variables omitted may have revealed additional critical issues beyond what is currently understood. The third limitation is the absence of Global Positioning System (GPS) coordinates from the crash locations. As a result, we were unable to map the blackspots of fatal motorcycle crashes nationwide. This constraint limits our ability to recommend location-specific safety measures, such as installing speed-calming infrastructure, improving signage, or enhancing lighting, and instead necessitates broader, region-level recommendations.

Additionally, the classification of rider errors, such as too fast (speeding), improper overtaking, and others, may have been influenced by crash severity and may not necessarily reflect the actual riding conditions. Police crash investigations in Ghana do not involve direct speed measurement; rather, assessments typically rely on eyewitness reports or post-crash inferences, which can introduce misclassification bias. Future research could explore age-period-cohort analysis, use vehicle registration as an exposure for regional trends, and, if such data are not available, use regional fuel consumption to normalise the trend.

Conclusion

This study highlights a significant and growing burden from motorcycle crashes in Ghana. The interplay between rider behaviours, temporal exposure, and infrastructural design creates a high-risk environment for motorcyclists. Risky behaviours, such as speeding and improper overtaking, significantly increase the hazards posed by road features, including sharp curves and unregulated intersections. These infrastructural vulnerabilities are further exacerbated during high-risk periods, such as nighttime, weekends or festival periods, when impaired visibility and alcohol use are more prevalent. Together, these interacting factors help explain the rise in fatal motorcycle crashes observed in Ghana. Addressing them through targeted, evidence-based interventions is essential to improving motorcycle safety and reducing the burden of traffic-related fatalities. The National Road Safety Strategy IV target of halving road fatalities will not be met, given the continued rise in fatal motorcycle crashes. The findings highlight the need for targeted interventions, such as rider training and licensing systems, enhanced infrastructure design, enforcement of traffic laws and speed management. Another plausible intervention is to promotion of vehicle safety technologies, including Anti-lock Braking Systems (ABS).

Abbreviations

ABS

Anti–lock Braking Systems

AIC

The Akaike Information Criterion

aRR

Adjusted Risk Ratio

BAC

Blood Alcohol Concentration

BRRI

Building and Road Research Institute

CI

Confidence Interval

cRR

Crude Risk Ratio

CSIR

Council for Scientific and Industrial Research

GPS

Global Positioning System

LMICs

Low-and-middle income countries

MAAP

Micro–computer Accident Analysis Package

SATPP

The Africa Transport Policy Program

TRL

Transport Research Laboratory

WHO

World Health Organisation

Authors’ contributions

**DG** : Conceptualisation, Methodology, Data curation, Formal analysis, Writing – original draft; **BNA** : Formal analysis, Writing – review and editing; **AM** , **DW & AG** : writing – review and editing; **DDJ** & **FA** : Validation, writing – review and editing; **JJ** : Conceptualisation, Methodology, Supervision, writing – original draft, writing – review and editing.

Funding

DG is supported by the University International Postgraduate Award (UIPA) from the University of New South Wales (UNSW). JJ is supported by the National Health and Medical Research Council (NHMRC) as an emerging leader 1.

Data availability

Data for this study will be made available upon request. The National crash data is publicly available at the Building and Road Research Institute, Kumasi.

Declarations

Ethics approval and consent to participate

Ethical approval was not obtained for this study. The study used secondary data compiled by the Building and Road Research Institute in Kumasi, based on national police crash files. The data is publicly available, accessible and anonymised. No human participants were directly involved, and the data does not contain any personally identifiable information.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.