Traumatic brain injury in Sierra Leone: Epidemiological burden and the urgent case for neurosurgical investment

Abstract

Introduction

Traumatic brain injury (TBI) represents a significant worldwide public health concern.

Research question

We evaluated the prevalence of TBI documented over 5 years at Connaught Teaching Hospital in Freetown, Sierra Leone.

Materials and methods

A retrospective chart review was conducted of all patients presenting with neurologic or neurosurgical conditions from January 2019 through December 2023. All available paper charts were reviewed. All patients diagnosed with traumatic brain injury were included in the present study.

Results

Between 2019 and 2023, Connaught Hospital evaluated 899 patients with TBI, the majority of whom were men between the ages of 21 and 40. The most common diagnosis was mild traumatic brain injury (45.3%) followed by severe head injury (16.0%). Road traffic accident was the most frequent cause of TBI (65.0%). Only twelve people had surgical treatment, none of which was neurosurgical. 48.8% of patients had a documented improvement in GCS at time of discharge, and there was a 11.4% in-hospital mortality rate.

Discussion and conclusion

Africa's trauma systems face numerous challenges, particularly neuro-related cases. The lack of specialists and equipment is a major problem in the region. TBI is very common in Sierra Leone, has a significant negative influence on young people's lives, and is comparable to that of other developing countries in the area. There is an urgent need for more research and systems development given the increasing burden of TBI on health resources in these nations.

Highlights

• •Traumatic brain injury incidence is high at Sierra Lone's primary tertiary referral hospital.
• •No neurosurgical interventions were performed over five years, regardless of injury severity or GCS score.
• •Road traffic accidents dominate as a mechanism of injury for TBI, identifying clear prevention priorities.

1. Introduction

An estimated 90% of the 4.7 million trauma-related deaths that occur each year take place in low- and middle-income countries (LMICs) and damage to the central nervous system is responsible for around half of trauma-related deaths in the world (Eaton et al., 2017). More than 10 million individuals suffer from traumatic brain injuries (TBIs) each year (National Academies of Sciences et al., 2022). TBI continues to be a public health concern, contributing to high rates of morbidity and mortality in the younger population even with efforts to enhance care. According to projections made by the World Health Organization, TBI ranked third in terms of death and disability in 2020, making it one of the most urgent and underappreciated public health issues globally (Dunne et al., 2020).

TBI is projected to affect up to 69 million individuals annually, with LMICs having the highest frequency, and the available data indicates that the death rate following a severe traumatic brain injury is twice as high in LMICs as it is in high-income (HICs) (Dewan et al., 2019; Naik et al., 2023). The annual incidence of TBI in low- and middle-income countries (LMICs) is 150–316 cases per 100,000 people. A growing middle class, urbanization, the availability of less expensive vehicles and motorcycles, and an aging and growing population are all contributing factors to the rising number of TBI deaths, particularly in the context of a poorly developed health care system in many of these regions (Eaton et al., 2017).

In addition, the mortality rate from falls and TBI rises with age from childhood to individuals over 75 years old (Miller et al., 2021), making it an increasing concern as life expectancy increases across the world.

The long-term deficits and disabilities brought on by TBI have a profound effect on families and society, costing the public a huge financial burden. Even in the most resource-rich environments, TBI is a heterogeneous disorder that can be challenging to manage and predict. It requires a variety of prehospital, acute treatment, and rehabilitation management options, making it a complex and multifaceted condition with effects that reach further than just the primary in-hospital care. Because of the enormous burden TBI places on individuals and families, as well as communities and economies due to the large unmet burden of disease, it is imperative to understand the current incidence, treatment patterns, and outcomes to identify systems gaps and urgent areas of improvement. [BLINDED FOR REVIEW] in Freetown, Sierra Leone is the primary tertiary referral center and only teaching hospital for surgical specialties in the country, with a catchment area of at least 5 million people. However, there have not been neurosurgical services at [BLINDED FOR REVIEW] or anywhere in the public sector during the study period. The purpose of this study was to document TBI data in Sierra Leone to inform necessary public health policy and resource-appropriate clinical management while the Ministry is working to develop a neurosurgical department, to help reduce TBI-related morbidity and mortality in the country.

2. Methods

A retrospective chart review was conducted of all patients presenting to [BLINDED FOR REVIEW] in Freetown, Sierra Leone with neurologic or neurosurgical conditions from January 2019 through December 2023. All paper charts housed in the records department, morgue, accidents and emergency department (A&E), inpatient ward books, and outpatient clinics were reviewed. Inpatient ward documentation was truncated with only name, age, dates, and diagnoses, so included limited data and contributed to data missingness of database. Inclusion criteria were broad in an attempt to capture any case that may have required a neurosurgical evaluation or neurosurgical care (e.g. evidence of spine or head injury after MVC or fall, enlarging head, back pain, weakness or hemiparesis, spina bifida, etc.). Any patient diagnosed with a traumatic brain injury, skull fracture, or intracranial hemorrhage after presentation due to a traumatic injury were included in the present study (138 of which were from inpatient ward books with limited information).

Variables collected include patient identification number, age, gender, profession, area of residence, date of symptom onset or injury, date of presentation, triage level in emergency department, admission status, service and level, neurologic exam and presenting Glascow Coma Scale (GCS) score, past medical history, brief history of presentation, other injuries, use of helmet or seatbelt if road traffic accident, use of drugs or alcohol if trauma, location of injury, prior treatment and location of treatment for presenting complaints, interventions performed, medications given, imaging ordered and completed, and discharge and follow-up data, if available. Data missing from the paper charts were left blank in the data collection form. If discharge date was not explicitly stated in the chart, it was listed as the last date of any documentation or note in the chart. All data was collected via Google Form and subsequently downloaded to an Excel spreadsheet.

Each patient was reviewed and reason for presentation, presenting symptoms, and diagnosis were categorized manually by a single author (M.E.H.S.) who is a senior neurosurgery trainee, familiar with the brief patient evaluation, triage, chart review process, and the clinical context in Sierra Leone. Because so few patients completed ordered imaging, most had presumptive diagnoses made based on the available documented clinical context, including history of presenting symptoms, medical history, and initial exam findings. The fidelity of the diagnosis and categorizations were attempted to be standardized as much as possible by interval validation during data collection and understanding the clinical context to improve interpretation of the limited available data.

Descriptive statistics were calculated using available-case analysis. For each variable, percentages were based on the number of patients with non-missing data, resulting in denominator variability between variables.

This study had research ethics approval from [BLINDED FOR REVIEW] (IRB202300606) and the [BLINDED FOR REVIEW] (004/07/2023) which supported data collection through July 2024. Patient consent was waive due to the retrospective nature of chart review.

3. Result

Of the 2992 patients in the neurosurgery database, eight hundred and ninety-nine patients were diagnosed with TBI with a 3:1 male to female ratio. The majority of TBI patients were between the ages of 21 and 40 (47.6%) (Table 1). Most patients were from within Freetown proper (within approx. 10 km from hospital) while fewer had a documented residence in other provinces. 80%% of the population had no prior medical history documented, whereas 4.2% had a history of hypertension.

Table 1.

Demographics of the patient population, presented as N(%). Percentages are representative of the total number of patients with data for each variable. Continuous variables presented as mean (median [IQR] range). KM-kilometers; HIV- Human immunodeficiency virus; TBI- traumatic brain injury.

	Number (N = 899)	Percentage
Sex (n = 898)
Female	218	24.3
Male	680	75.7
Age (n = 811)
	27.6 (26.0 [17,36] 0-87)	mean (median [IQR] range)
≤13 Years	169	20.8
14-20 Years	105	12.9
21-40 Years	386	47.6
41-60 Years	107	13.2
>60 Years	44	5.4
Distance from Residence to Connaught (n = 715)
	26.1 (5.1 [5.1,12.2] 1-360)	mean (median [IQR] range)
∼5 Km	54	7.6
5-10 Km	442	61.8
10-50 Km	155	21.7
>50 Km	64	9.0
Past Medical History (n = 899)
Asthma	3	0.33
HIV	1	0.11
TBI	1	0.11
Diabetes	4	0.44
Hypertension	38	4.2
Seizure	1	0.11
Stroke	5	0.56
Drinks Alcohol	5	0.56
Smokes Tobacco	3	0.33
None	719	80.0

A little less than half (49%) of the patients had been evaluated at other hospitals or medical facilities prior to being transferred to [BLINDED FOR REVIEW] (Table 2). Altered mental state (AMS) was the most prevalent symptom upon presentation, followed by intact patients and then headache. Road traffic accidents accounted for 65% of presentations, followed by falls from height (14.3%). A majority of patients with initial GCS on first evaluation had mild TBI (GCS documented as 13-15 (45.3%), though a significant percentage of patients were missing documented GCS upon arrival. Most patients were ranked as yellow or orange in triage (42.1%, 31.8%, respectively), indicating a moderate severity and urgency of injury upon first evaluation. A majority of patients (81.6%) were admitted to the hospital, most commonly to the emergency department (46.8%) and orthopedic surgery service (27.5%).

Table 2.

Documented findings of the first evaluation of the patients diagnosed with TBI, presented as N(%). Percentages are representative of the total number of patients with data for each variable. Continuous variables presented as mean (median [IQR]). EMS = Emergency medical services; AMS= Altered mental status; GCS = Glascow Coma Scale; A&E = Accidents and Emergencies unit; ICU= Intensive care unit.

	Number (N = 899)	Percentage
Time from Symptom Onset to Presentation (n=523)
Up to 1 day	429	82.0
2-7 days	62	11.9
8-14 days	12	2.3
3-4 weeks	5	0.96
1-6 months	9	1.7
7-12 months	2	0.38
>1 year	4	0.76
Previous Treatment (n = 761)
Yes	373	49.0
No	388	51.0
Previous Treatment Location (n = 367)
Home	2	0.54
Ambulance/EMS	27	7.4
District Hospital	67	18.3
Missionary Hospital	51	13.9
Private Hospital/Clinic/Pharmacy	34	9.3
Government Hospital	186	50.7
Triage Color (n = 660)
Green	81	12.3
Yellow	278	42.1
Orange	210	31.8
Red	91	13.8
Reason for Presentation (n = 761)
Altered Consciousness	39	5.1
Assault	67	8.8
Fall from Height	109	14.3
Fall from Standing	12	1.6
Road Traffic Accident	495	65.0
Seizure	1	0.13
Unspecified Trauma	31	4.1
Weakness	7	0.92
GCS on Arrival (n = 626)
	12.0 (14.0 [9,15])	mean (median [IQR])
History of Loss of Consciousness (n = 761)
Yes	637	83.7
No	124	16.3
History of Seizure-Like Activity (n = 761)
Yes	58	7.6
No	703	92.4
Symptoms Upon Presentation (n = 700)
AMS	338	48.3
Aphasia	7	1.0
Back Pain	8	1.1
Craniofacial Injury	17	2.4
Depressed Skull Fracture	12	1.7
Facial Palsy	5	0.71
Headache	49	7.0
Hemiparesis	17	2.4
Hypertensive	10	1.4
Lethargy	4	0.57
Neck Pain	48	6.9
Paraparesis	1	0.14
Quadriparesis	9	1.3
Seizure	22	3.1
Sluggish Pupil	13	1.9
Weakness	13	1.9
Fixed Pupil	9	1.3
Numbness	1	0.14
Intact	221	31.6
Admitted to Hospital (n = 756)
Yes	617	81.6
No	139	18.4
Service of Admission (n = 628)
A&E	294	46.8
Medicine	50	8.0
Orthopedics	173	27.5
Pediatrics	46	7.3
Surgery	65	10.4
Level of Care of Admission (n = 616)
A&E	331	53.7
Ward	251	40.7
ICU	34	5.5

82.5% of patients had X-rays ordered but only 29.1% completed them (Table 3). Out of the 270 requests, only 76 patients (28.1%) had CT scans performed, leading to only 13.4% of neurologic diagnosis being confirmed on imaging with the remainder of patients being treated presumptively based on clinical evaluation and mechanism of injury. Only twelvepatients (1.6%) had surgery performed by the orthopedic surgery service for polytrauma.

Table 3.

Patterns of in-hospital care of the patients diagnosed with TBI, presented as N(%). Percentages are representative of the total number of patients with data for each variable.

	Number (N = 899)	Percentage
Diagnosis (n = 899)
Degloving Injury	9	1.0
Depressed Skull Fracture	18	2.0
Open Skull Fracture	5	0.56
Facial Fractures	23	2.6
Orbital Fracture	1	0.11
Mild Head Injury	407	45.3
Moderate Head Injury	129	14.3
Severe Head Injury	144	16.0
Unspecified Traumatic Brain Injury	128	14.2
Intracranial Hemorrhage	18	2.0
Subdural Hematoma	23	2.6
Subarachnoid Hemorrhage	18	2.0
Seizure	13	1.4
Spinal Cord Injury	12	1.3
Spine Fracture	2	0.22
Unspecified Spine Injury	16	1.8
Orthopedic Fracture	69	7.7
X-Ray Ordered (n = 761)
Yes	628	82.5
No	133	17.5
X-Ray Completed, of those ordered (n = 628)
Yes	183	29.1
No	445	70.9
CT Ordered (n = 761)
Yes	270	35.5
No	491	64.5
CT Completed, of those ordered (n = 270)
Yes	76	28.1
No	194	71.9
MRI Ordered (n = 761)
Yes	0	0
Neurologic Diagnosis Confirmed with Imaging (n = 890)
Yes	119	13.4
No	780	87.6
Surgery, non-neurosurgical (n = 751)
Yes	12	1.6
No	739	98.4
Length of Hospital Stay (n = 858)
	6.7 (3.0 [1,8] 0-164)

There was a 11.4% documented in-hospital mortality, while a majority of patients were discharged with an undocumented final neurologic status and GCS score (Table 4). Just 8.8% of patients followed up in outpatient clinic after being discharged.

Table 4.

Documented findings upon discharge of patients diagnosed with TBI, presented as N(%). Percentages are representative of the total number of patients with data for each variable. GCS = Glascow Coma Scale; A&E = Accidents and Emergencies unit; ICU= Intensive care unit.

	Number (N = 899)	Percentage
GCS on Discharge (n = 371)
GCS ≤8	5	1.3
GCS 9-12	5	1.3
GCS 13-15	361	97.3
GCS Difference from Admission (n = 371)
Improved	181	48.8
Stable	187	50.4
Worse	3	0.81
Mortality (n = 760)
Yes	87	11.4
Level of Care on Discharge (n = 645)
A&E	341	52.9
Ward	278	43.1
ICU	26	4.0
Discharge Location (n = 694)
Home	606	87.3
Home Against Medical Advice	64	9.2
Another Hospital	18	2.6
Care Facility	3	0.43
Presented for Follow-up (n = 751)
Yes	66	8.8

4. Discussion

Traumatic brain injury is a major global source of avoidable morbidity and mortality, yet it is disproportionately concentrated in regions where surgical capacity to address it is most limited. This review identified a high incidence of TBI at the primary referral hospital in Sierra Leone, none of which underwent neurosurgical intervention, regardless of the severity of injury. This finding is itself the central argument of this paper: understanding who is sustaining TBI, how they are arriving, and what happens to them before and after presentation is the necessary foundation upon which any surgical intervention program must be built.

TBI has been dubbed a "silent epidemic" with cases frequently going unreported by patients or unrecognized by medical personnel (Tegegne et al., 2023). It's silence is compounded in LMICs by a lack of systematic data collection, making epidemiological studies such as this one foundational, rather than supplemental, to clinical progress. Due to the immediate costs of medical care and rehabilitation as well as the intangible costs associated with impairment and lost productivity, particularly in the young population who tend to be the primary providers for families, TBI has a significant negative societal and economic impact-one that falls hardest on national economies already under strain (Righy et al., 2025). Poverty and limited health literacy contribute to the 3–4 times higher TBI-related mortality rate in LMICs compared to HICs (Naik et al., 2023). TBI often affects people in their prime working years, resulting in substantial losses through absenteeism, decreased productivity, and permanent job loss. In LMICs, the combined effects of these home microeconomic factors result in more significant macroeconomic difficulties for national economies.

Road traffic accidents (RTAs) were the most common reason for presentation in this cohort, consistent with the broader literature on TBI in LMICs (Rosenberg et al., 2020; Barki et al., 2023). This pattern reflects systemic vulnerabilities such as rapid motorization without commensurate investment in safety infrastructure, weak enforcement of traffic laws, and inadequate protection for vulnerable road users, including pedestrians and two-wheeler riders (Barki et al., 2023; Ramesh et al., 2014). Additionally, enforcement of helmet use continues to be a primary safety concern in many LMICs, frequently falling between 0% and 45% in nations in Asia and Africa (Abdi et al., 2022). While helmet use was not a variable in the present study due to lack of documentation in the chart, this has been anecdotally noted to be a frequent problem in the country by the study team and a primary area of “low hanging fruit” for TBI prevention programs in the country. These upstream factors are not merely contextual background-they define the preventable faction of TBI burden and should directly inform public health programming alongside clinical capacity building.

In addition to the economic effects, TBI is stigmatized in LMICs, which makes healing and social reintegration even more difficult. The strain on impacted people and their families can be exacerbated by stigma, which can result in social isolation, weakened support networks, and less prospects for employment and rehabilitation. Furthermore, TBI survivors frequently struggle to obtain appropriate treatment and rehabilitation due to a lack of culturally relevant neurological care resources and general awareness in many LMICs, which prolongs long-term disability and dependency (Ramesh et al., 2014). Addressing stigma and improving healthcare literacy must therefore be considered an integral component of any comprehensive TBI care strategy.

Our results showed that 46.9% of TBI patients received pre-arrival evaluation or care, indicating that a significant proportion of the population may have required primary stabilization or interventions prior to evaluation at the tertiary trauma center. This is a vital part of the TBI care pathway as pre-hospital care can be integral for stabilization in trauma and prevention of secondary brain injury (Rakhit et al., 2021; Spaite et al., 2019). However, the majority of patients were driven to the hospital in private vehicles with only a few arriving in ambulances. This situation is not unusual in LMICs, as other research has indicated that cabs are typically the most popular way to get to the hospital, particularly following an RTC (Meena et al., 2018). Similarly, post-injury follow up rates in this cohort were strikingly low, representing a substantial gap in the continuum of care. Since follow-up care has been demonstrated to positively impact long-term healing, symptom management, and improvements in quality of life following TBI, low follow-up rates are alarming. Enhancing education and stressing the value of clinical follow-up may contribute to higher rates and, eventually, better recovery paths for TBI patients.

Future projections of TBI will be necessary to evaluate the demand for long-term care, assisted living facilities, and the psychological expenses associated with neurocognitive impairment brought on by TBI (Eaton et al., 2017). When evidence-based management protocols are implemented, all TBI patients should receive uniform multidisciplinary care based on guidelines. This will address the standard care practices, which include timely placement of interventions like neck collars, appropriate radiographic studies, referral of the patient to the appropriate level of care, and frequent neurological checks to prompt further interventions as necessary. This can be done even in a setting with limited resources with the appropriate training and resource allocation and is being implemented at Connaught Hospital through a team of multidisciplinary teaching efforts in the emergency department, wards, and ICU by an international team working to support the newly developing neurosurgery department at the hospital.

5. Limitations

Due to many lost, unreadable, and inconsistent documents in the hospital's records room, detailed information on TBI is still scarce. Accurate epidemiological analysis is hampered by manual chart reviews, which frequently come across incomplete patient files, faded handwriting, and inconsistent recording standards among staff. These problems, which are prevalent in LMIC settings with inadequate resources, lead to underreporting of TBI incidence, causes (such as traffic accidents), and outcomes, which compromise the accuracy of prevalence estimates. The inability to identify long-term consequences due to a lack of post-discharge follow-up limits our investigation. Additionally, because of presentation bias at Connaught and other hospitals, we are unable to fully estimate the incidence of TBI in Sierra Leone. Patients who passed away before coming to the Connaught Hospital or who only visited the district and other hospitals other than Connaught were not included in our count. The population location, cultural variances, socioeconomic status, and time of the study all affect the epidemiologic results. Small patient numbers in the bulk of the examined articles provide a less accurate picture of a region as populous as sub-Saharan Africa.

6. Conclusion

This study reflects the high incidence of TBI in Sierra Leone, though likely underreports, especially in the most severe of cases. The absence of neurosurgical intervention should not be interpreted as evidence of low surgical need, but rather a direct reflection of the absent surgical capacity, likely leading to significantly worse outcomes. Epidemiologic data that outlines the commonest mechanisms of injury, severity distribution, pre-hospital pathways, and care gaps are not ancillary to surgical program development; rather, they are its prerequisite. This study provides that foundational profile for Sierra Leone and supports the ongoing multidisciplinary international efforts to support the development of the first neurosurgery department in the public sector, with a first neurotrauma-trained surgeon introduced to [BLINEDED FOR REVIEW], in early 2025.

These findings were used to develop evidence-based, multi-disciplinary management protocols adapted for the resource-limited environment, focusing on spinal stabilization, rapid neurologic examination and triage, and more defined admission pathways, as well as daily care checklists. Such protocols are implementable even where operative neurosurgery is yet unavailable, and their adoption creates the baseline into which surgical capacity is being introduced as it grows. Ongoing work at [BLINDED FOR REVIEW] exemplifies this staged approach. Standardized clinical data collection at sites like this will also be essential for future projections of TBI burden, long-term care demand, and the true economic cost of neurologic impairment in this population.

Author contribution statement

MSK, KSO, and MEHS were responsible for project ideation, and initiation. A. Sannoh, AB, AMB, A. Kalokoh, A. Sall, and MEHS participated in team recruitment in Sierra Leone, formation of data collection forms, organization, and team oversight. A. Sankoh, A. Sall, Alusine Barrie, Alhassan Barrie, AMB, IKT, FMACCJ, SB, ECE, EUP, A. Kamara, A. Kalokoh, HM, A. Sannoh, and MEHS participated in data collection. MEHS was responsible for data cleanup and analysis. Alusine Barrie, Ahassan Barrie, and MEHS were responsible for manuscript composition. AQ-H, MSK, KSO, and MEHS were responsible for manuscript revisions and finalization. All authors had the opportunity to review and approve the final draft of the manuscript.

Ethical approval and informed consent statements

This study has research ethics approval from the University of Florida IRB (IRB202300606) and the Office of the Sierra Leone Ethics and Scientific Review Committee (004/07/2023) which supports data collection through July 18, 2024. Patient informed consent requirements were waived due to the retrospective chart review nature of the study.

Funding statement

This work was funded by the Mission:BRAIN Lisa Hannegan Humanitarian International Neuroscience Scholarship and the United States Fulbright Program which were received by Dr. Still. No other authors received funding for this study.

Declaration of competing interests

The authors declare no conflicts of interest or disclosures.