Time delays in emergency stroke care in a low-resource referral hospital in Ghana

Abstract

Background

Patients in sub-Saharan Africa face significant delays in receiving appropriate stroke care, which negatively impacts outcomes. This study aimed to quantify the time delays in acute stroke care at the Komfo Anokye Teaching Hospital (KATH) and determine the effect on mortality.

Methods

This was a secondary analysis of prospectively collected data on patients ≥ 18 years with Computed Tomography (CT) scan-confirmed stroke treated at KATH’s adult emergency department (ED) from November 2021 to March 2022. Patients were initially enrolled in a pilot stroke registry by trained research assistants, who documented relevant time points in their care. Patient outcome (dead or alive) was determined at time of hospital discharge and three months post-discharge. Data was analyzed using STATA™ version 16. The median times from stroke onset to ED arrival, physician evaluation, CT scan imaging, and treatment were measured and the relationship with mortality determined.

Results

Eighty-six patients with confirmed stroke were analyzed, comprising 40 males and 46 females. Ages ranged between 29 and 86 years, with mean of 57.4 years (SD 14.3). The median time from stroke onset to arrival at KATH ED was 35.3 h (IQR: 12.3–79.5). The median time from ED arrival to first physician evaluation was 1.3 h (IQR: 0.5–2.6); to CT imaging was 14.1 h (IQR: 4.3–40.8); and to antiplatelet treatment (for ischemic stroke) was 31.1 h (IQR: 16.1–42.5). The cumulative mortality rates at three months post-discharge were 8.7 % for patients who arrived at KATH’s ED within 4 h of symptom onset, 43.5 % for those arriving between 4 and 24 h, and 47.8 % for those arriving after 24 h, p = 0.036.

Conclusion

Significant delays occurred in all stages of stroke care at KATH’s ED. Improving stroke education and implementing contextually appropriate stroke codes can enable early patient presentations, improve intervention times, and reduce mortality rates.

African Relevance

• Stroke outcomes are better when patients receive timely treatment. This requires implementation of strategies and guidelines that can aid early symptom recognition, facilitate rapid transportation to appropriate care facilities, and ensure prompt investigation and administration of therapeutic agents. Unfortunately, such strategies or guidelines are rarely utilized or adhered to in Sub-Saharan Africa.

• This study provides local data quantifying the delays and highlighting the challenges in achieving timely treatment for stroke in a resource-constrained setting in Ghana, as well as identifying the key areas for interventions which would be the most impactful.

• This study identified considerable time delays in all stages of the stroke care process, including delays in hospital arrival after onset of stroke symptoms and intra-facility delays that led to late evaluation, imaging and treatment.

• This study showed that the post-hospital discharge mortality rates for stroke patients were significantly higher in patients who experienced longer prehospital delays. This suggests that by educating the general public and primary care providers on how to identify stroke symptoms and steps to take in seeking immediate and appropriate care, patient outcomes could be improved.

• In addition to promoting stroke educational campaigns, implementing stroke codes and protocols that prioritize the early evaluation, imaging and treatment of stroke patients in healthcare facilities can reduce delays in our setting.

Alt-text: Unlabelled box

Background

Acute stroke is a time-sensitive disease requiring that affected patients seek specialist medical attention within minutes of symptom onset [1]. Early diagnosis, with appropriate and timely treatment, can help patients recover from this disease with significant return of neurological function [1]. In light of this, most high-income countries with advanced health systems routinely carry out educational and public awareness campaigns on early recognition of stroke symptoms [2,3]. At the same time, protocols and guidelines to facilitate immediate transportation and interventions have been developed and are regularly implemented [4].

The American Heart Association (AHA) and the American Stroke Association (ASA) have together published guidelines for stroke care, including recommended timelines for pertinent assessments and interventions [5]. Additionally, several easy-to-use tools like ACT FAST have been developed to aid the general public and prehospital care personnel recognize and act on stroke symptoms early [6]. The stroke chain of survival further emphasizes early transportation to a stroke-capable health facility after symptoms have been identified [7]. Such facilities should ideally have stroke care teams and prioritize early assessment, imaging and treatment of affected patients, including services for thrombolysis and rehabilitation [7]. In low- and middle-income countries (LMICs), stroke education is generally lacking, and well-structured systems and policies for comprehensive stroke care are currently inadequate, resulting in many stroke patients being diagnosed late and expectedly, morbidity and mortality outcomes are poorer [8,9].

The recommended door-to-imaging time for stroke patients who arrive at a treatment facility is 45 min [10]. For patients who meet eligibility, stroke onset to thrombolysis time should be <4.5 h; to endovascular clot retrieval <24 h, and to antiplatelet therapy, should be within 24 to 48 h [11]. Globally, these timelines are hardly met, but delays are more significant in LMICs, particularly within sub-Saharan Africa [8,9,12].

At the Komfo Anokye Teaching Hospital (KATH) in Ghana, stroke thrombolysis services are available. However, this intervention is significantly underutilized, likely due to late patient presentations. Additionally, the lack of local protocols and guidelines, as well as the absence of a multidisciplinary team to coordinate the care of stroke patients in this facility, may be contributing to delays in receiving appropriate interventions. Currently, there is limited data on the timeliness of stroke diagnosis, areas in the care chain where delays occur, and how delays affect patient outcomes.

This study aimed to map the pathway for acute stroke care in the adult emergency department (ED) at KATH and quantify the time delays from onset of stroke symptoms to the point of receiving definitive treatment. Secondarily, it sought to determine the association between these time delays and mortality. This information can highlight specific areas for recommendations and improvements that can optimize care and reduce morbidity and mortality from strokes.

Methodology

Study design

This was a secondary analysis of prospectively collected data on 86 patients with Computed Tomography (CT) scan-confirmed stroke who were captured in a pilot stroke registry at KATH’s adult ED between November 2021 and March 2022.

Setting

KATH is the second-largest tertiary hospital in Ghana, located in Kumasi city, with an estimated population of 3.3 million [13]. KATH houses a 37-bed adult emergency unit with 24-hour access to laboratory and radiology services, including a CT scan, which is the primary imaging modality for acute strokes at this facility. KATH has a dedicated stroke ward staffed by specialist neurologists and is the major referral center for stroke patients in the region. Patients admitted through the adult emergency department are triaged using the South African Triage Scale (SATS) [14].

Participants

Participants included in this registry were individuals aged ≥ 18 years who were either referred or self-reported to the KATH ED with clinical features or a CT scan suggestive of stroke. Patients were excluded if they refused consent, died on arrival, or if their symptoms were secondary to a traumatic event or non-vascular in origin (for example, hypoglycemia, toxic/metabolic encephalopathies, alcohol withdrawal, cerebral infections, or endocrine emergencies).

Data collection

Triage nurses identified consecutive patients who met inclusion criteria upon arrival and notified the research team. Following initial resuscitation and informed consent, trained research assistants collected data for the registry, which included demographic information, relevant past medical histories, referral status, mode of transportation, clinical features, brain imaging reports, as well as the date and time stroke symptoms started, date and time of hospital arrival, date and time of first physician evaluation, date and time of brain imaging, and the date and time of treatment with antiplatelet, anticoagulant and/or thrombolytic agents if applicable. Patients were followed till discharge and then up to 3 months post-discharge, and their status (alive or dead) was determined. Patient enrolment occurred 24 h a day, and data was mainly obtained through patient or caregiver interviews and direct observation of care procedures. All dates and time points were corroborated with timestamps on patients’ electronic medical records and image files. Status at 3 months post-discharge was determined through scheduled follow-up phone calls. Data was entered electronically on tablet computers using the Research Electronic Data Captures (REDCap™) software [15].

Statistical procedures

Data analysis was performed using STATA™ version 16.0 (StataCorp. 2019. Stata Statistical Software: Release 16. College Station, TX: StataCorp LP). Descriptive analyses were presented using frequencies and percentages The median times for the measured time intervals were determined with corresponding interquartile ranges (IQR). Using Chi-square tests and Fisher exact tests where necessary, the relationships between categorized time intervals and mortality were tested. All statistically significant results were set at ≤ 0.05.

Ethics

This study was entirely voluntary, and written consent was obtained from patients or their legal guardians prior to enrolment. Ethics approval was obtained from the KATH IRB. (KATH IRB/AP/089/21).

Results

A total of 110 patients with suspected stroke were recruited into the pilot registry. Eleven patients could not perform CT imaging, 8 patients had alternative diagnoses on CT scans, and 5 had normal CT scan findings, and thus were excluded. The remaining 86 patients with CT scan-confirmed stroke were analyzed. They comprised 40 (46.5 %) males and 46 (53.5 %) females. Ages ranged from 29 to 86 years with a mean of 57.4 (SD 14.3) years. Seventy-four patients (86.1 %) had at least basic-level education (first 6 years of primary formal education), and 63 (73.3 %) had stable employment. The majority of patients (73, 84.9 %) were referred from lower-level health facilities, 44 (51.2 %) arrived via private vehicles or taxis and 42 (48.8 %) came via ambulance. From CT scan reports, 40 patients (46.5 %) had ischemic stroke and 46 (53.5 %) had intracerebral hemorrhage (ICH). The hospital mortality rate was 46.5 % (Ci: 0.356–0.575), and the cumulative mortality rate at 3 months post-discharge was 53.5 % (Ci:0.424–0.643) (Table 1).

Table 1.

Demographic features, prehospital information and outcomes of stroke patients treated at the Komfo Anokye Teaching Hospital Emergency Department (KATH ED).

Variable	Frequency N=86	Percentage ( %)
Age < 40 years	9	10.47
Age: 40–60 years	45	52.33
Age > 60 years	32	37.20
Male	40	46.51
Female	46	53.48
Triage category: Red	30	34.88
Triage category: Orange	56	65.12
Employed	63	73.26
Unemployed	23	26.74
No formal education	12	13.95
Primary-level education	37	43.02
Secondary-level education	28	32.56
Tertiary-level education	9	10.47
Referred	73	84.88
Not referred	13	15.12
Arrived via ambulance	42	48.84
Arrived via taxi/private vehicle	44	51.16
CT scan done at referring hospital	38	44.19
CT scan done at KATH	48	55.81
Ischemic Stroke on CT scan	40	46.51
Intracerebral Hemorrhage on CT scan	46	53.49
Discharged alive	46	53.49
Discharged dead	40	46.51
Alive at 3 months post-discharge	40	46.51
Dead within 3 months post-discharge	6	6.98
Cumulative mortality (3 months post-discharge)	46	53.49

CT: Computed Tomography; KATH: Komfo Anokye Teaching Hospital.

The median time from onset of stroke symptoms to arrival at KATH ED for all patients was 35.3 h (IQR: 12.3–79.5). Between the two stroke types, the median arrival time for patients with ischemic stroke (31.8 h) was shorter than that of those with ICH (44.6 h). For a sub-group of 13 (15 %) patients who identified the KATH ED as the first health facility they visited after stroke symptoms started (labeled as non-referrals), the median time from symptom onset to ED arrival was 26.3 h (IQR: 8.1–55). The median time from arrival at KATH ED to first physician evaluation was 1.3 h (IQR: 0.5–2.6). For patients with ischemic stroke alone, it was 1 hour, and for those with only ICH, 1.7 h. Forty-eight (55.8 %) patients had their brain CT scans performed at KATH. Of these, the median time from arrival to getting CT imaging was 14.1 h (IQR: 4.3–40.8). For patients who performed CT imaging at KATH and were diagnosed with ischemic stroke, only 14 (16.3 %) had complete data to calculate the time from arrival to treatment with antiplatelet agents, which was 31.1 h (IQR: 16.1–42.5). When all the patients with ischemic stroke (both referral CT scan-diagnosed and KATH CT scan-diagnosed) were analyzed together, the median time from ED arrival to antiplatelet therapy was 8.0 h (IQR: 4.15–18.48) (Table 2).

Table 2.

Time intervals in the care pathway for stroke patients treated at Komfo Anokye Teaching Hospital Emergency Department (KATH ED).

Time Interval	Total			Ischemic stroke only			Intracerebral hemorrhage only
	N	Median time (IQR)	Missing data	N	Median time (IQR)	Missing data	N	Median Time (IQR)	Missing data
Symptom onset to arrival at ED (total)	86	35.29 (12.25–79.5)	–	40	31.81 (10.24–70.88)	–	46	44.63 (12.5–82.9)	–
Symptom onset to arrival at ED (non-referrals)	13	26.33 (8.08–55)	–	11	26.33 (8.08–55)	–	2	43.75 (2.12–85.38)	–
ED arrival to first physician evaluation	86	1.28 (0.5–2.6)	–	40	1.03 (0.46–1.88)	–	46	1.7 (0.67–2.98)	–
Symptom onset to CT imaging at referring hospital	6	8.22 (7.13–15.78)	32	5	8.27 (8.18–15.78)	–	1	5.08 (5.08–5.08)	–
ED arrival to CT imaging at KATH	48	14.14 (4.35–40.83)	–	29	13.42 (4.52–33.1)	–	19	14.87 (3.92–66.27)	–
ED arrival to antiplatelet (CT done @ KATH alone)				14	31.12 (16.07–42.52)	15
ED arrival to antiplatelet (CT @ KATH + CT @ referral hospital)				18	7.49 (3.93–18.22)	22

ED: Emergency Department; CT: Computed Tomography.

The cumulative mortality rates at 3 months post-discharge were 8.7 % for patients who arrived at KATH’s ED <4 h after symptom onset, 43.5 % for those who arrived between 4 and 24 h, and 47.8 % for those arriving after 24 h, p = 0.036 (Table 3).

Table 3.

Relationship between categorized time intervals (delays) and mortality at 3 months post-discharge.

Categorized time interval	Total ( %)	Alive	Dead	X2/Fisher exact test	P value
Symptom onset - ED arrival (total)	N = 86			0.033	0.036*
< 4 h	6 (6.98)	2 (5.00)	4 (8.70)
4–24 h	28 (32.56)	8 (20.00)	20 (43.48)
>24 h	52 (60.47)	30(75.00)	22 (47.83)
Symptoms onset–ED arrival (non-referrals)	N = 13			0.755	0.585
< 4 h	2 (15.38)	1 (50.00)	1 (50.00)
4- 24 h	4 (30.77)	1 (25.00)	3 (75.00)
>24 h	7 (53.85)	4 (57.14)	3 (42.86)
ED arrival - physician evaluation	N = 86			0.795	0.672
< 30min	14 (16.28)	5 (35.71)	9 (64.29)
30–60min	12 (13.95)	6 (50.00)	6 (50.00)
>60min	60 (69.77)	29(48.33)	31(51.67)
ED arrival to KATH imaging	N = 48			0.570	0.512
≤ 30min	0
0.5–4 h	14 (29.17)	6 (42.86)	8 (57.14)
>4–8 h	6 (12.5)	1 (16.67)	5 (83.33)
>8 h	28 (58.33)	9 (32.14)	19(67.86)
ED arrival to antiplatelet therapy (CT scan done at KATH)	N = 14			0.685	0.377
< 4 h	1 (7.14)	1 (100)	0
4–24 h	3 (21.43)	1 (33.33)	2 (66.67)
>24hours	10 (71.43)	7 (70.00)	3 (30.00)

X²: Chi-Squared test, KATH: Komfo Anokye Teaching Hospital, ED: Emergency Department, CT: Computed tomography.

Discussion

Stroke continues to be a major health concern in Ghana, affecting a growing number of individuals across the region [16]. Our study results showed that individuals aged between 40 and 60 years were the most affected, and the majority of strokes identified were intracerebral hemorrhages (ICHs). This aligned with findings in an earlier research by Sarfo et al. in 2018 [17]. Notably, over 80 % of affected patients had received basic formal education, consistent with the country’s overall basic education rate of 71 % [18]. A significant majority of patients initially sought care from primary health facilities that lacked capabilities for advanced stroke care.

The time from onset of stroke symptoms to hospital arrival differs across several African countries, influenced by the strength of health systems and resources [12,19,20]. On average, a median time of 31 h has been reported [8]. In this study, the time to hospital arrival was 35.3 h, with approximately 93 % of patients reporting after 4 h of symptom onset and nearly two-thirds coming after 24 h. This was comparable to studies in Uganda and DR Congo, where 9 out of every 10 stroke patients arrived at a health facility after 3 h and 4.5 h, respectively [21,22]. These time intervals are less than ideal and remarkably higher than observed in developed countries within Europe and North America. In these regions, patients reported to a hospital within 3 to 6 h after stroke onset [[23], [24], [25]].

There are multiple reasons for prolonged prehospital delays in LMICs. First, the general lack of education and public awareness on how to recognize stroke symptoms has been highlighted in several African publications [8,[26], [27], [28]]. Additionally, there are no guidelines outlining the measures to take or the steps to follow after a stroke occurs [8,9]. In contrast, countries in Europe and North America have widespread public educational campaigns on stroke, which have significantly altered perspectives and enhanced early stroke identification and presentation to care facilities [2,3]. Implementing similar public awareness campaigns coupled with feasible, cost-effective innovations, such as Mobile Stroke Units, have the potential to reduce prehospital delays in stroke care within LMICs [29,30].

Stroke education should ideally be extended to health professionals in primary care facilities who may be seeing a significant proportion of cases. Aside from being equipped to recognize stroke symptoms promptly, primary care providers must assess their capabilities and the capabilities of their facility to provide the required care and make immediate referrals when necessary.

Given that the majority of stroke patients in this study were referred from lower-level health facilities, it is reasonable to assume that the delayed arrivals observed may have been partly due to delays in the decision to refer. This assumption is supported by previous research [31,32]. It is also likely that the challenges of ambulance transportation, such as limited availability, poor road infrastructure, and high service costs, which are more prevalent in LMICs, may contribute to delayed stroke referrals from primary care facilities, as supported by previous literature [8,27,29].

In high-income countries with advanced prehospital systems, ambulances transport patients directly from their homes and workplaces. In these settings, trained paramedics can identify stroke symptoms promptly, facilitate timely activation of stroke codes and ensure transport to stroke-capable facilities [6,7]. A study by Mosley et al. [33], in Melbourne, Australia, assessing the impact of ambulance use in acute stroke care, found that early identification of stroke symptoms by paramedics and prompt ambulance transportation significantly reduced arrival times for affected patients.

Our data also revealed that patients with ischemic stroke arrived at the KATH ED sooner than those with ICH. This contradicted previous research, which suggested that ICHs often presented with more severe and acute symptoms, prompting patients or their caregivers to seek medical attention sooner [34,35]. However, the clinical features of ICHs, which include headaches, vomiting, altered consciousness, focal neurologic deficits or seizures, are nonspecific and some patients or primary care providers may not immediately associate them with strokes [36]. This can lead to delays in hospital presentation as some individuals may seek alternative treatments, or delays in primary care diagnosis leading to delayed referrals [22,37]. Additionally, patients with large ICH who are often critically ill may require ambulance transportation between health facilities, further hindering timely referral [8].

Regarding patient outcomes, this study determined that longer pre-hospital delays were associated with increased mortality. Similarly, Garcia Ruiz et al. [38] observed poorer outcomes in stroke patients with longer prehospital delays in Spain. Also, a study by Denti et al. [39], in Italy found that stroke patients who arrived after two hours of symptom onset had higher mortality compared to those who arrived earlier. In contrast, Turan et al. [40] suggested that early hospital arrivals were mostly seen in patients with higher stroke severity, which may be accounting for the high mortality.

The stroke chain of survival emphasizes early evaluation upon hospital arrival, prompt brain imaging, and immediate administration of therapeutic agents to ensure favorable outcomes [7]. However, multiple intra-facility delays make these targets unachievable, particularly in LMICs. In this study, the median time from hospital arrival to first physician evaluation was 1.3 h and nearly 2 h for those with ICH alone. This delay is likely due to triage teams not being equipped to identify stroke symptoms or appreciate the seriousness of this disease. As a result, evaluations by emergency physicians or stroke teams could occur several hours after patients have reported [41]. Additionally, overcrowding in emergency rooms, with patients waiting long hours to access care, can lead to stroke patients getting lost in the pool, hindering timely evaluation [42]. With clear stroke protocols in place, triage teams can be guided to identify stroke patients early and facilitate prompt physician evaluation [32,41]. Also, having protocols that advocate for advance notice from paramedics and referring hospitals can help receiving facilities adequately prepare and plan for stroke patients, potentially reducing the time to critical assessments and evaluation [43].

A non-contrast brain CT scan is a recommended initial imaging modality for stroke [4]. The AHA and ASA guidelines recommend that stroke patients receive neuroimaging within 25 min of emergency department arrival and images interpreted within 20 min to determine stroke subtype and eligibility for treatment interventions like thrombolysis [10]. In this study, the median time from arrival to obtaining a head CT scan was 14.1 h, a significant departure from this target. The time to brain imaging for stroke patients differs based on the strength of health systems and resources. In Burkina Faso, stroke patients had neuroimaging after 21 h of hospital arrival, while in Johannesburg, South Africa, the door-to-imaging time was approximately 8 h [12,19]. High-income countries in Europe and North America may also experience delays in meeting the recommended timelines, but the delays are not as severe. For instance, data from 4 inner-city hospitals in Berlin, Germany revealed that it took stroke patients an average of 108 min to get neuroimaging after arriving in the emergency room [44].

Even when a CT scan is available, imaging for stroke patients in countries like Ghana may still be delayed for several reasons. The high cost of imaging and the fact that patients have to pay before imaging are the major barriers driving this delay. Additionally, the lack of institutional protocols prioritizing early imaging for stroke patients, the distance of CT scan facilities from the emergency department, and staff limitations are other reasons documented in literature [12,19,44]. Addressing these issues, in addition to promoting effective collaboration among a multidisciplinary team of emergency care providers, including triage nurses, physicians, porters and radiologists, can significantly speed up the process of getting brain imaging for stroke patients [19].

This study also measured the time interval from hospital arrival to vital treatments and interventions like antiplatelet administration. For patients with confirmed ischemic stroke, who underwent CT imaging at KATH, this time interval was 31 h. Patients who arrived from a referring hospital and had already completed CT imaging were expected to start this treatment sooner. As such, when analyzing all patients with ischemic stroke, including those with referral CT scans, the time from ED arrival to antiplatelet treatment was remarkably shorter. However, if the prehospital delays observed in this study are factored in, it becomes evident that the total time from stroke onset to antiplatelet therapy exceeds the recommended time window proposed by the AHA and ASA [5]. Furthermore, the time-to-treatment observed in this study significantly exceeded the time window for other interventions like thrombolytic therapy [5]. This, amongst other reasons, explains the abysmally low rate of stroke thrombolysis performed at this facility [9]. Other contributing factors which have been highlighted in previous research include the high cost or unavailability of thrombolytic agents, overcrowded EDs with limited staffing, and the lack of education or local protocols to guide this treatment [9,45,46].

Limitations

This was an observational single-center study, that focused solely of patients with CT scan-confirmed strokes. As such, the results cannot be generalized to the entire population. Additionally, there was a substantial amount of missing data from referring hospitals and treatment logs, which impacted the calculation of some of our time intervals, such as the time to referral CT imaging and the time to initiating antiplatelet therapy. It is also likely that patients with severe disease were not stable enough for transport to the CT room, which may have impacted time to neuroimaging. However, the absence of stroke severity scores, which were not routinely documented at the KATH ED, limits our ability to draw this conclusion. Furthermore, we did not account for stroke severity and other potential confounders, like the presence of infections and social support after discharge, when calculating mortality outcomes.

Conclusion

Significant delays were observed in all stages of stroke care for patients treated at the KATH ED, and mortality rates were higher among patients who experienced longer prehospital delays. Enhancing stroke education for public and primary care providers can help reduce patient arrival times. Additionally, implementing contextually appropriate stroke codes and protocols that prioritize timely patient evaluation, investigation and treatment are highly recommended, especially in busy, overcrowded and resource-constrained EDs across Sub-Saharan Africa.

Dissemination of results

The findings of this study were presented to a global audience at the 6th African Conference of Emergency Medicine (AfCEM 2022) held in Accra, Ghana. The study abstract was also accepted for an oral presentation at a Medical Knowledge Fiesta organized by the Ghana College of Physicians and Surgeons in Accra in 2023. Additionally, during a Departmental meeting at the emergency department of the Komfo Anokye Teaching Hospital, where the study was conducted, we shared the results and recommendations of this study with the management and clinical team.

Funding

This study was funded by the National Institute of Neurological Diseases and Stroke (NINDS) of the National Institute of Health (NIH), USA, as part of the Northern Pacific Global Health Fogarty Fellowship Program, 2001/2022 cohort.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the author used Grammarly and Word Tune in order to improve readability and language. After using this tool, the author reviewed and edited the content as needed and takes full responsibility for the content of the publication.

CRediT authorship contribution statement

Hussein A Yakubu: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Writing – original draft. Richmond O Marfo: Investigation, Software, Formal analysis. Jonathan Boakye-Yiadom: Methodology, Investigation, Formal analysis. Freda M Aidoo: Investigation, Project administration. Fred S Sarfo: Supervision, Writing – review & editing. Rockefeller A Oteng: Conceptualization, Methodology, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declared no conflicts of interest.