Road Traffic Accidents (RTAs) and injuries pose a significant threat to both public health and development. The incidence of road traffic accidents is higher in low- and middle-income nations such as Kenya than in high-income nations. As the eighth biggest global cause of mortality, about 50 million individuals are injured and over 1.3 million people lose their lives in traffic-related accidents each year [1, 2]. Every year, 17 road fatalities per 100,000 people are reported worldwide. More than 75% of RTA casualties occur in the 15 to 44-year-old economically active demographic group, making RTAs the second highest cause of death in this age group [3, 4]. The estimated economic damage from RTAs in several nations is as high as 3% of their gross domestic product [4].
The anticipated rate of road fatalities in Kenya is 20.9 per 100,000 people, which is greater than the rate in the European region (10.3 per 100,000 people) [35]. In 2010, the Kenya traffic police reported a total of 3,055 road traffic fatalities, and around 7% of them were motorcycle riders [37].
Tragedies like this are common in Kenya. On April 1, 2024, a five-vehicle car crash on the Nairobi-Mombasa highway killed 11 people, seven of whom were from the same family returning from Easter holidays and many of whom were children. In fact, road traffic crashes are the leading cause of death for young adults and the 12th most common cause of death across all age groups.
A descriptive analysis of road traffic accident (RTA) and injury data in Kenya was done using routine accident reports, official statistical abstracts, published and unpublished surveys. The numbers killed increased by 578%, while non-fatal casualties rose by 506% between 1962 and 1992. Fatality rate per 10,000 vehicles increased from 50.7 to 64.2, while fatality per 100,000 population ranged between 7.3 and 8.6. 66% of the accidents occurred during daytime. 60% of the reported RTAs occurred on rural roads and had a higher case fatality rate (CFR) of 16% compared to those occurring in urban areas (11%). Human factors were responsible for 85% of all causes. Vehicle-pedestrian collisions were most severe and had the highest CFR of 24%, while only 12% of injuries resulting from vehicle-vehicle accidents were fatal. Utility vehicles, 'matatus' and buses were involved in 62% of the injury producing accidents. Of all traffic fatalities reported, pedestrians comprised 42%, passengers 38%, drivers 12%, and cyclists 8%.
Data Collection and Analysis
Data from the recent Kenyan demographic and health survey completed in 2022 were evaluated in this study. A weighted sample of 145,880 household members provided the data. Enumeration areas served as the primary sampling units, and households served as the secondary sampling units in a stratified two-stage cluster sampling procedure. Key causes of road traffic accidents in Kenya were investigated using univariate and multivariable multilevel logistic regressions. The prevalence of road traffic accidents was determined to be 6.33% (95% CI = 0.32-12.07). The prevalence of road traffic accidents in Kenyan household members was considerably lower when compared to other preceding findings.
Read also: Waste to Wonder: Flip Flop Art
The 2022 Kenyan Demographic and Health Survey (2022 KDHS) was conducted by the Kenya National Bureau of Statistics (KNBS) in collaboration with the Ministry of Health (MoH) and other partners. This survey is the seventh iteration of the KDHS. A cross-sectional study design was employed, with data collected from February 17 to July 31, 2022. The Demographic and Health Surveys (DHS) Program, funded by the United States Agency for International Development (USAID), provides technical and financial support for health and population surveys worldwide. This study utilized data from the 2022 Kenya Demographic and Health Survey (KDHS), a nationwide community-based cross-sectional survey.
The Kenya Household Master Sample Frame (K-HMSF) was used as the sampling framework for the 2022 KDHS. This framework was developed using data from the 2019 Kenya Population and Housing Census (KPHC), which identified 129,067 enumeration areas (EAs). A total of 10,000 EAs were selected for the K-HMSF through probability proportional to size. These EAs were divided into four subsamples, and based on the required sample size, either a single subsample or a combination of subsamples was used. The 2022 KDHS sample was drawn from subsample one of the K-HMSF. Our source population included all household members in Kenya, while our study population included those household members who had reported any road traffic-related accidents (RTAs) among their members in the 12 months preceding the survey, within the designated enumeration areas (EAs) or primary sample units of the survey clusters [40].
The present study utilized data from the 2022 Kenya Demographic and Health Survey (KDHS), a nationally representative, community-based cross-sectional survey. The sample for the 2022 KDHS was based on a calculated size of 42,300 households with a cluster size of 25 households. The KDHS employed a two-stage stratified sampling design. In the first stage, 1,692 clusters were selected from the Kenya Household Master Sample Frame (K-HMSF) using the Equal Probability Selection Method (EPSEM), with clusters randomly chosen within each sampling stratum. In the second stage, a household listing was performed for each selected cluster, and 25 households were randomly selected from each of the clusters. In order to sample 42,022 households for the 2022 KDHS, this procedure was carried out. Only the pre-selected clusters and households were used for interviews; replacement of these units was not permitted during data collection [38].
The study focused on the occurrence of any form of RTAs or injuries among household members in the 12 months prior to the survey as the primary outcome variable. Household members were asked the question “How many members were injured in a road accident in the last 12 months?” to determine whether there were any injured family members. The independent variables included demographic factors such as the participants’ age (< 25 years, 25-34 years, 35-44 years, > 44 years), sex (male/female), and marital status (not married, married, divorced/widowed). Behavioral characteristics like current smoking status (yes/no) and smoking in the last 24 h (yes/no) were also assessed. Socioeconomic indicators such as education level (not educated, primary, secondary and above), household wealth index (poor, middle, rich), and household size (1-5 members, 6-10 members, > 10 members) were included. The analysis also considered access to media and technology, including internet usage (yes/no), television viewing (yes/no), and radio listening (yes/no). Community-level factors, such as media exposure (low/high), education level (low/high), and wealth (low/high) were taken into account. Lastly, the ownership of means of transportation, including bicycles, motorcycles, and trucks (yes/no for each), was examined as an independent variable.
To ensure the highest quality DHS data, the research team will take meticulous steps. They have thoroughly trained the data collectors, supervisors, and field editors, and provide ongoing supervision to maintain data integrity. Standardized questionnaires, translated into local languages, was used consistently, with data processing specialists carefully managing the entry and processing to uphold accuracy. Before the full-scale data collection, a pre-test was conducted, followed by a debriefing session to gather feedback and refine the questionnaires as needed. It’s important to note that the DHS is a nationally representative survey carried out every five years by trained professionals, collecting data from households, women, men, biomarkers, and health institutions. The data source for this investigation was the most recent DHS data for Kenya. To ensure a comprehensive dataset with a large sample size representative of the population, the standard DHS data set was utilized [39].
Read also: Discover Sentrim Elementaita Lodge
The STATA formatted DHS files for the person records (PR) were obtained. The data was then accessed, cleaned, coded, and combined to create the necessary variables for the analysis. Prior to statistical analysis, the data were weighted using the sample weight to account for the probability sampling design and non-response, in order to restore the representativeness of the sample. Statistical software such as STATA version 17 and Microsoft Excel 2019 were utilized to conduct both descriptive and analytical analyses. Due to the hierarchical structure of the DHS data, where household members were nested within clusters, the assumptions of the standard logistic regression model may not hold true. This led to the fitting of a multilevel binary logistic regression using four models. To assess the variation in RTAs across the clusters, the null model (Model 1) was utilized without including any predictor variables. The second model (Model 2) included variables at the individual level. The third model (Model 3) incorporated variables at the community level. In the final model (Model 4), both individual and community-level variables were simultaneously fitted to examine their association with the prevalence of RTAs. The log-likelihood and deviance tests were used to compare the models, and the best-fitted model was determined to be the one with the highest log-likelihood and the lowest deviance value. Furthermore, the multilevel modeling approach served as a litmus test to determine whether a multi-level or conventional logistic regression should be utilized, thereby justifying the employment of such a framework.
The multilevel model was assessed using the log-likelihood ratio test (LLR), median odds ratio (MOR), intra-class correlation coefficient (ICC), and proportional change of variance (PCV). According to the null model, there were significant differences in RTAs status between clusters (p-value = 0.001; 2u0 = 0.072). The ICC was 49.04% (95% CI: 47.89,50.86), indicating that unobserved factors at the community or individual level or differences between communities accounted for 49.04% of the overall variability in RTAs. This suggests that a multilevel logistic regression model is preferable to a single-level logistic regression model for obtaining reliable results [41].
This survey comprised weighted samples from 145,880 household members whose personal records were included. In terms of age, approximately 69,483 (47.63%) of the study participants were 44 years old or over. Regarding place of residence and marital status, the majority, 96,652 (66.25%), lived in rural areas.
Factors Contributing to RTAs
Any RTA may result in various types of injury. Injury is characterized as physical harm that happens when a human body is purposely or unintentionally exposed to unacceptable harm caused by thermal, mechanical, electrical, or chemical energy, or the lack of necessary heat or oxygen [4, 7]. Injuries harm society because they account for 10% of fatalities and 16% of the global burden of disease [8, 9]. The risk of RTAs is greater in productive age groups [8, 10]. According to the WHO Global Burden of Disease (GBD) study, injuries are responsible for 9% of all mortality and 12% of the total burden of disease globally [11]. The WHO report also indicated that low and middle-income countries account for more than 90% of injury cases. Of these, 21% of injuries occur in sub-Saharan African countries [11]. The highest annual rate of RTA deaths is in the continent of Africa, where it is 27 per 100,000 people [12]. Due to the continent’s continued economic expansion and rise in motorization, the issue of road traffic accidents in Africa may potentially worsen over the next few decades [13, 14]. Similarly, research from South Africa and Zimbabwe has revealed that injuries account for the most significant deaths and morbidities [15]. Additionally, injuries caused by traffic collisions, drowning, poisoning, falls, burns, and violence are a significant and growing burden in East African countries [9, 16]. However, in developed countries, injury rates have declined by 30.9%, with an annualized rate of reduction of 1.6% [17]. The significant causes of injury are road traffic accidents, firearms, drowning, falls, burns, and poisoning [18].
Determinants of RTAs involve socioeconomic, demographic, and behavioral factors. Studies have linked factors like gender, age, income, occupation, and household size to RTA risk in Qatar, and Nigeria [21-23]. Behavioral factors are crucial too. Experts have evaluated interventions to reduce alcohol-impaired driving in Thailand and examined risk perceptions in Qatar [21, 22, 24]. They’ve also identified individual, family, and social influences on RTAs among young people. Broader systemic factors have been analyzed as well. Meta-analyses have revealed socioeconomic, demographic, and traffic-related influences on accident frequency [25]. Assessments of the injury burden in Africa and disparities in developed countries, compared to African cities, have provided important insights [26-28].
Read also: Best Nairobi Excursions
Furthermore, previous studies have revealed that several determinants contribute to RTAs among household members. These include social determinants of health such as education, income, rural/urban settlement, and marital status have been associated with RTAs [30, 31]. Similarly, the external environment, including traffic environment and road environment, can contribute to RTAs. Factors such as the increased number of vehicles, growing traffic density, and the types of vehicles on the road are positively associated with a higher number of crashes, injuries, and deaths [32, 33]. Factors like inexperience, lack of driving skills, risk-taking behaviors, irritability, and anger can contribute to RTAs [22, 34]. Factors specific to the vehicle, such as the condition of the vehicle, can also contribute to RTAs [22, 33]. Other factors such as substance use have also been identified as risk factors for RTAs [22].
According to a 2009 assessment from the Kenyan government, motorcycle riders now account for more accidents on the road than matatu (refer to privately owned minibuses that function as shared taxis or public transportation. and “matatu drivers”) would therefore be the drivers of these particular vehicles [36].
Although injuries are largely preventable, they continue to be a widespread health problem in Kenya. However, there are not many comprehensive studies on the epidemiology of vehicle-related accidents in the country. This could be a result of a dearth of national or sub-regional trauma registries, with statistics on injuries typically coming from examinations of individual hospital records. To constructively engage public officials and develop effective policies for injury and accident prevention, more research must be conducted in this area.
Leveraging Data for Road Safety
The World Bank has been working for years in the Kenyan context to develop the AI tools to detect and georeference road car crashes. The intention has been to find the most treacherous tracks of road and focus road safety efforts on them. The reason for doing this is simple: years of road safety recommendations have not enabled countries to make their investments matter.
Road networks are extensive, road safety investments expensive, and information scarce. Data can make a difference. Data can help identify high-risk corridors where crashes are concentrated, allowing policy makers to target the few kilometers of road with the highest crash risk. Moreover, data can be used to evaluate investments and trigger course corrections.
While official data can be scarce, new sources of information can be leveraged to obtain good data. Indeed, 92% of crash fatalities occur in low- and middle-income countries (LMICs) that typically lack digital systems for recording crashes. If at all, crashes are recorded on paper and can underestimate crashes. But times have changed. Today, bystanders report crashes on social media in large numbers. With some ingenuity, the World Bank team was able to demonstrate that one can turn tweets into a publicly available dataset. In the case of Nairobi, this resulted in a dataset and map of over 30,000 geocoded crashes.
The World Bank Smart and Safe Kenya Transport (smarTTrans) team focused on Kenya because at a rate of 28 road traffic fatalities (RTF) per 100,000, the country exemplifies the tragedy taking place across sub-Saharan Africa (27 RTF/100,000). The dataset represents crashes from time and locations where bystanders are more likely to see and report crashes. Moreover, more visible crashes, such as crashes that result in traffic delays, may be more likely to be reported by bystanders irrespective of their severity. Posts do not reliably capture information on fatalities or injuries. Some posts reference that a crash resulted in injuries or fatalities; however, not all posts contain this information and the ability of bystanders to accurately ascertain injury information is uncertain.
The algorithm to geolocate crashes is not perfect. We develop and implement an algorithm that geolocates crashes based on the text of the post, using references to landmarks and roads. To test the accuracy of the algorithm, we manually code the locations of 1 year of posts. The algorithm determines the correct location for 65% of crashes from the truth dataset (recall). Similarly, in March 2020, there was a sharp reduction in crashes after social distancing measures, including a curfew, were implemented in response to COVID-19. This reduction could result from a reduction in crashes, which has been reported in other contexts-but could also result from less users on the road to report crashes.
Despite these limitations, the data can still be useful to identify high-risk locations in the city. To map high risk crash locations, we group crashes within 500 meters of each other into clusters-where clusters with the most crashes could be considered blackspots. The map and corresponding table below show that crashes are spatially concentrated.
Adventures in Crowdsourcing: Using Crowdsourced Data for Traveler Information
Our aspiration for this dataset is to fully harness the potential of crowdsourced efforts. While the dataset is generated from crowdsourced reports, by making the dataset public we hope to crowdsource analysis of the data.
Popular articles:
tags: #Kenya
