Abstract
Background
Births that are small-for-gestational-age (SGA) represent a major cause of neonatal morbidity and mortality, particularly in sub-Saharan Africa. The contribution of specific placental histopathological lesions and their association with maternal risk factors remain poorly characterized in this high-burden setting. This study aimed to determine the incidence of adverse neonatal outcomes in SGA births and to identify associated placental and maternal predictors in northern Uganda.
Methods
A prospective cohort study was conducted at Lira Regional Referral and Kiryandongo General Hospitals in Uganda from 15th April 2025 to 15th August 2025. A total of 151 women delivering singleton SGA infants were enrolled. Maternal sociodemographic, clinical, and obstetric data were collected via a structured questionnaire, and placentas were examined using standardized histopathological protocols. The primary outcome was a composite of adverse neonatal events, including admission to the neonatal intensive care unit (NICU), Apgar score <7 at 5 minutes, preterm birth, or early neonatal death. Multivariable Poisson regression was used to identify independent predictors of the composite outcome.
Results
The incidence of adverse neonatal outcomes was 48.3% (95% CI: 40.4–56.4). Placental abnormalities were identified in 47.7% of cases, with inflammatory lesions (8.6%), vascular lesions (7.9%), and villous infarctions (6.0%) being the most prevalent findings. After adjusting for confounders, the presence of any placental abnormality was the strongest independent predictor of adverse outcomes (adjusted Relative Risk = 2.23; 95% CI: 1.45–3.43; p < 0.001). Maternal obesity (Body Mass Index ≥30 kg/m2; aRR = 1.61; 95% CI: 1.06–2.46; p = 0.027) and higher maternal income (>200,000 Ugandan Shillings; aRR = 1.66; 95% CI: 1.03–2.69; p = 0.037) were also significantly associated with an increased risk of adverse outcomes.
Conclusion
Nearly half of SGA neonates in this northern Ugandan cohort experienced an adverse neonatal outcome. Placental abnormalities were common and were associated with adverse outcomes, and maternal obesity was identified as a potentially modifiable risk factor. These findings support the value of structured placental examination after SGA deliveries for etiologic classification and risk stratification and highlight the need for further research to determine whether targeted antenatal interventions addressing maternal nutrition and weight management can reduce adverse neonatal outcomes.
Keywords
1. Introduction
A birth that is small-for-gestational-age (SGA), defined as birth weight below the 10th percentile for gestational age and sex, is a major determinant of perinatal health worldwide.1,2 An estimated 17% of live births are SGA globally, with the greatest burden in low and middle-income countries (LMICs), where the majority of cases occur.3,4 While SGA prevalence in high-income settings ranges from 5–7%,5,6 rates in South Asia and sub-Saharan Africa often exceed 20%. 3 SGA infants face increased risks of neonatal complications, including perinatal asphyxia, hypoglycemia, hypothermia, NICU admission, and neonatal death.7,8
Sub-Saharan Africa carries the highest neonatal mortality rates globally, and SGA is an important contributor.9,10 In East Africa, an individual participant-level meta-analysis showed that SGA infants had approximately double the odds of neonatal death compared with appropriately grown infants; risk rose markedly when SGA co-occurred with prematurity. 11 In Uganda, neonatal mortality remains high (18.4 deaths per 1,000 live births), above the Sustainable Development Goal target (<12/1,000). 12 Studies from rural Uganda have reported SGA prevalence up to 21.4%. 13 Despite this burden, there is limited local evidence describing why some SGA newborns experience severe early adverse outcomes while others do not, particularly regarding the role of placental pathology and modifiable maternal factors.
Placental insufficiency is a central biological pathway underlying fetal growth restriction and adverse perinatal outcomes.14,15 The placenta mediates nutrient and oxygen transfer and supports fetal growth; therefore, structural or functional impairment can plausibly translate into neonatal compromise. 16 Histopathological examination of the placenta can reveal specific lesions that provide a morphological basis for its dysfunction and key lesions associated with FGR include those of maternal vascular malperfusion (MVM), such as villous infarctions, decidual vasculopathy, and accelerated villous maturation, which are indicative of uteroplacental ischemia. 17 Studies from high-income settings report that such pathological findings are present in up to 84% of placentas from SGA births. 18 However, there is a scarcity of comprehensive data on the prevalence and types of placental lesions in SGA pregnancies from sub-Saharan Africa, limiting the understanding of the underlying mechanisms in the region with the greatest disease burden.
Maternal clinical and environmental exposures can shape placental health and may modify risk among SGA births. In sub-Saharan Africa, infections such as HIV, malaria, and syphilis are prevalent and can contribute to placental inflammation and impaired function.19–21 These exposures are associated with inflammatory placental lesions (e.g., chorioamnionitis, villitis, and intervillositis) that may disrupt placental architecture and nutrient transfer.22,23 At the same time, many LMICs including Uganda are experiencing a nutritional transition with rising maternal overweight and obesity.24,25 Maternal obesity can promote a pro-inflammatory and metabolically dysregulated intrauterine environment, plausibly contributing to placental vascular dysfunction, impaired transport capacity, and adverse perinatal outcomes.26,27 This metabolic dysregulation is mechanistically linked to placental endothelial dysfunction, impaired nutrient transport, and the development of MVM lesions, thereby increasing the risk of adverse perinatal outcomes.28,29 Other maternal factors associated with SGA in East Africa include young maternal age, primiparity, and delayed initiation of antenatal care. 30
While SGA is a well-established driver of adverse neonatal outcomes in Uganda, the specific contributions of placental pathology and the interplay of maternal risk factors in this high-burden setting remain poorly characterized. Understanding these relationships is essential for developing effective risk-stratification strategies and targeted interventions. We hypothesized that histopathological placental abnormalities are associated with a higher risk of adverse early neonatal outcomes among SGA births, independent of maternal clinical risk factors. Therefore, this study aimed to: 1) determine the incidence of a composite of adverse neonatal outcomes among SGA births at two referral hospitals in northern Uganda; 2) describe the prevalence and types of placental histological abnormalities in this cohort; and 3) identify the independent maternal and placental predictors of these adverse outcomes.
2. Methods
2.1. Study design, setting, and population
A prospective cohort study was conducted in the maternity units of Lira Regional Referral Hospital (LRRH) and Kiryandongo General Hospital (KGH) in northern Uganda from 15th April 2025 to 15th August 2025. This was a short-term prospective cohort: participants were enrolled before delivery and neonates were followed through delivery and up to 7 days postpartum to ascertain early neonatal outcomes, including early neonatal death. Both study sites are regional referral facilities receiving high-risk obstetric referrals, including hypertensive disorders of pregnancy, suspected infections, and medically complex pregnancies from lower-level units. The catchment area has endemic infectious disease burden (e.g., malaria and HIV), which can influence maternal and placental pathology. Lira Regional Referral Hospital has NICU capacity and access to pathology services, while Kiryandongo General Hospital provides comprehensive emergency obstetric and neonatal care with established referral pathways for critically ill newborns.
LRRH is a 401-bed public referral facility serving a catchment population of approximately 2.5 million, equipped with a neonatal intensive care unit (NICU) and pathology services. KGH is a 100-bed general hospital also providing comprehensive obstetric and neonatal care. The study population comprised pregnant women admitted in the latent phase of labor who subsequently delivered a singleton SGA infant and provided written informed consent. Women with multiple gestations, uncertain gestational age, obstructed labor, or those unable to provide consent were excluded.
2.2. Sample size and sampling
The sample size was estimated to determine the incidence of adverse neonatal outcomes among SGA births with acceptable precision. Assuming a prevalence of 50%, 95% confidence level, 5% margin of error, finite population correction for approximately 250 SGA deliveries in the prior year, and 5% non-response, the minimum required sample size was 151. This was distributed proportionally between Lira Regional Referral Hospital (81) and Kiryandongo General Hospital (70). Eligible participants were enrolled consecutively until the required sample size was attained. Multivariable regression was conducted as a secondary exploratory analysis.
2.3. Data collection
Upon enrolment, trained research assistants administered a structured researcher-developed questionnaire to obtain sociodemographic information, including maternal age, level of education, marital status, occupation, and self-reported monthly household income, together with selected maternal medical and obstetric history variables. The questionnaire was developed based on the study objectives and review of related literature. Prior to the main study, the tool was pilot-tested among approximately 10% of the estimated study population at a similar health facility that was not included in the final study. The pilot exercise helped improve the clarity, wording, consistency, and overall flow of the questionnaire before actual data collection. However, the questionnaire was not formally validated using psychometric validation methods.
Additional clinical and obstetric information was extracted from patient records using a standardized data abstraction form. These records included antenatal care (ANC) cards and handheld records for documented blood pressure measurements, infections and treatments received, and diabetes screening results where available. Maternity inpatient charts and partographs were reviewed for intrapartum events and mode of delivery, while laboratory records provided information on malaria testing and HIV/syphilis results where documented. Neonatal charts and NICU registers were also reviewed to obtain information on NICU admissions, indications for admission, and early neonatal outcomes.
Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m2) using either measurements taken at admission in labor or the most recent ANC-recorded measurements where available. Since pre-pregnancy BMI measurements were not routinely available, BMI estimates were interpreted as late-pregnancy or peripartum BMI, which may have introduced some degree of misclassification.
Maternal clinical factors assessed included hypertensive disorders of pregnancy such as pre-eclampsia/eclampsia, diabetes in pregnancy, suspected or confirmed infections, premature rupture of membranes, antepartum haemorrhage, and other clinician-documented pregnancy complications selected based on biological plausibility and previous evidence linking them to SGA births and adverse neonatal outcomes.
2.4. Placental collection and histopathological examination
Immediately following delivery, placentas were collected and placed in containers with 10% buffered formalin. Specimens were labelled with a unique study identifier and transported to the Department of Pathology at Mulago National Referral Hospital in Kampala. There, a certified pathologist, who was blinded to all clinical and neonatal outcome data, performed a standardized gross and microscopic examination. Tissues were processed through dehydration, clearing, and embedding in paraffin wax. Sections of 3–5 mu were cut, mounted on glass slides, and stained with Hematoxylin and Eosin (H&E). Lesions were classified according to established criteria, including features of MVM (e.g., infarction, accelerated villous maturation), fetal vascular malperfusion (FVM), and inflammatory patterns (e.g., villitis, chorioamnionitis).
Placental lesions were classified at the lesion level during pathology review using standardized criteria. However, for inferential analysis, lesion-level subcategories were not modeled separately because several categories were infrequent and were not available in a sufficiently consistent analyzable format across all case-level records for stable multivariable estimation. Therefore, lesion-specific findings are presented descriptively, whereas the regression analysis used a binary variable indicating the presence of any significant placental abnormality.
2.5. Study variables
The primary outcome was a composite adverse neonatal outcome, defined as the occurrence of one or more of the following: NICU admission, Apgar score <7 at 5 minutes, preterm birth (<37 completed weeks), or early neonatal death (within 7 days of birth). The primary explanatory variable was the presence of any significant placental abnormality on histopathology. Other variables included maternal age, BMI (≤24.9, 25–29.9, ≥30 kg/m2), education, monthly income, history of hypertension or diabetes, and documented infections during the index pregnancy. No qualitative data were collected. The composite outcome was chosen to capture serious early neonatal compromise in this referral-hospital setting. Its individual components were also reported separately.
Sociodemographic data were obtained by questionnaire. Maternal clinical variables were abstracted from ANC cards, maternity charts, partographs, and laboratory records where available. Neonatal outcomes and NICU admission data were obtained from neonatal charts and NICU registers. Maternal infection was categorized as malaria, UTI, STI, HIV, clinical chorioamnionitis, or other documented infection. SGA was defined as birth weight below the 10th centile for gestational age.31,32
Placental abnormality was treated as the primary explanatory/prognostic variable rather than as an upstream maternal risk factor, because it represents a proximate pathological marker that may lie on the causal pathway between maternal exposures and neonatal compromise.
Monthly household income was categorized as ≤200,000 versus >200,000 UGX. This cutoff was selected a priori to reflect the Ugandan income context and to provide an interpretable dichotomy, given a reported national median monthly household income of about UGX 190,000. 33
2.6. Quality control
To ensure data accuracy and reliability, quality control measures were implemented at both the data collection and placental histopathology stages. During questionnaire administration, trained research assistants delivered questions in simple, clear language and minimized technical terms to promote accurate responses. To enhance accessibility, the questionnaire was translated into the local language and back-translated to maintain linguistic precision. For participants who did not speak English, the local-language version was read aloud to ensure full comprehension. Immediately after each interview, the principal investigator reviewed the completed questionnaires and abstraction forms for completeness and internal consistency, reducing the risk of missing or inconsistent data. In addition, all data collection tools were reviewed daily, and selected variables were cross-checked against source documents (ANC cards, maternity records, laboratory records, and neonatal/NICU registers) where available.
For placental assessment, standardized operating procedures were followed for specimen handling, processing, slide preparation, and lesion classification, consistent with established consensus criteria. The laboratory/pathology personnel involved received targeted orientation, and routine checks were performed to ensure adherence to protocols. To strengthen diagnostic reliability, 10% of placental slides were randomly selected and independently re-examined by a second pathologist at an external laboratory (Lancet Laboratories Uganda), providing an additional layer of quality assurance. Inter-rater agreement was high, and any discrepancies were resolved through consensus review.
2.7. Statistical analysis
Data were entered into Microsoft Excel 2019 and exported to STATA version 15.0 (StataCorp, College Station, TX, USA) for analysis. Descriptive statistics (frequencies and percentages for categorical variables; means with standard deviations for continuous variables) were used to summarize participant characteristics. The incidence of the composite adverse neonatal outcome and the prevalence of placental abnormalities were calculated with 95% confidence intervals (CIs). Because the outcome was common, Poisson regression with robust error variance was used to estimate crude (cRR) and adjusted (aRR) relative risks with corresponding 95% CIs.
For multivariable modeling, variables were selected using a combination of biologic plausibility and bivariate screening. Specifically, the following variables were forced into the multivariable model a priori as clinically/biologically relevant predictors and potential confounders of the placental abnormality–outcome relationship: maternal age category, BMI category, history of hypertension, history of diabetes, and infection status during the current pregnancy, together with the primary exposure (placental abnormality). In addition, employment status and household income category were included as a priori socioeconomic confounders. Other variables with bivariate p<0.20 were considered for inclusion and retained if they meaningfully altered effect estimates or improved model fit. A p-value <0.05 was considered statistically significant. Multicollinearity was assessed using the variance inflation factor (VIF), with values <10 indicating no concerning collinearity.
2.8. Ethical considerations
Ethical approval for this study was obtained from the Research Ethics Committee of Kampala International University (KIU-REC-2025-849) and clearance was granted by the Uganda National Council for Science and Technology. Administrative permission was also obtained from Lira Regional Referral Hospital and Kiryandongo General Hospital before commencement of the study. Written informed consent was obtained from all participants before enrolment into the study. For participants who could not read or write, the information sheet and consent form were explained in their preferred language by a trained research assistant. Sufficient time was provided for participants to ask questions and make an informed decision about participation. Those who agreed to participate signed or provided a thumbprint on the consent form in the presence of an independent literate witness.
3. Results
A total of 167 mothers were approached and assessed for eligibility. Of these, 158 were eligible, and 7 declined to participate, leaving 151 mothers with SGA deliveries enrolled in the study. All enrolled participants were included in the final analysis, as no records were excluded. The flow of participants through the study is shown in Figure 1. Participant screening and enrollment flow diagram.
3.1. Basic characteristics of the study participants
Sociodemographic characteristics of the study participants (N=151).
UGX= Ugandan shillings, LRRH = Lira Regional Referral Hospital; KGH = Kiryandongo General Hospital; UGX = Ugandan Shillings.
3.2. Incidence of adverse neonatal outcomes among SGA deliveries at LRRH and KGH in Uganda (N=151)
Of the 151 SGA neonates, 73 experienced an adverse outcome, resulting in an overall incidence of 48.3% (95% CI: 40.4% - 56.4%). However, 78 (56.4%) did not. Among the 73 neonates who suffered adverse outcomes, the most frequently observed outcome was admission to the Neonatal Intensive Care Unit (NICU), which occurred in 89.0% of cases. This was followed by low Apgar scores (69.9%), preterm birth (23.3%), and early neonatal death (6.8%) (Figure 2). The most common specific findings included inflammation (8.6%), vascular lesions (7.9%), and placental infarction (6.0%) (Figure 3). Bar Chart representing the patterns of Adverse Neonatal Outcomes N=73. Bar Chart representing the Common placental abnormalities.
3.3. Factors associated with adverse neonatal outcomes among SGA deliveries at LRRH and KGH
Bivariate analysis of factors associated with adverse neonatal outcomes.
Ref: Reference, UGX Ugandan Shillings, cRR crude relative Risk, CI confidence Interval, aRR adjusted relative risk.
Multivariate Poisson regression analysis of factors associated with adverse neonatal outcomes.
Ref: Reference, UGX Ugandan Shillings, cRR: crude relative Risk, CI: confidence Interval, aRR adjusted relative risk.
3.4. Multicollinearity assessment
An assessment for Multicollinearity was conducted for the variables included in the final multivariate model using the Variance Inflation Factor (VIF). The results indicated that there was no significant Multicollinearity among the predictor variables, as all VIF values were well below the common threshold of 10. The mean VIF for the model was 1.69, confirming the absence of collinearity issues and the stability of the model estimates.
4. Discussion
This study provides critical insights into the burden and drivers of adverse outcomes among SGA neonates in a low-resource, high-morbidity setting in northern Uganda. The principal findings are threefold: first, the incidence of adverse neonatal outcomes is remarkably high, affecting nearly one in two SGA births. Second, placental abnormalities are highly prevalent and represent the strongest independent predictor of these poor outcomes. Third, maternal obesity emerged as a significant, modifiable risk factor, while higher maternal income was paradoxically associated with increased risk.
The observed incidence of 48.3% for the composite adverse neonatal outcome is substantially higher than the 9–20% rates typically reported in high-income countries 34 but is consistent with the 40–60% range estimated for LMICs. 35 This finding aligns with recent data from Eastern Uganda, where composite adverse outcomes were documented in 48.7% of newborns of prehypertensive mothers and 40% of those born after an emergency obstetric referral, highlighting the vulnerability of high-risk pregnancies in the region.36,37
A crucial aspect of this finding is the dominance of NICU admission (89.0%) within the composite outcome. This reflects not only the high level of neonatal morbidity but also the functional capacity of the two referral hospitals to provide specialized care. In many lower-level facilities across sub-Saharan Africa without NICU capabilities, a significant proportion of these morbidities would likely progress to mortality. 38 The high observed incidence, therefore, signals a potential epidemiological shift in well-equipped referral centers from a primary burden of neonatal mortality to one of severe morbidity, a methodological consideration that complicates direct comparisons with studies from less-resourced settings.
The finding that placental abnormalities were the strongest predictor of adverse neonatal outcomes (aRR = 2.23) reinforces the placenta’s central role as the final common pathway through which maternal and environmental insults are transmitted to the fetus. 39 The multivariable analysis demonstrated that the effects of upstream risk factors, such as maternal hypertension and infections, were attenuated after accounting for placental pathology. This suggests that these maternal conditions exert their detrimental effects primarily by inducing placental damage, which in turn leads to fetal compromise. This causal sequence underscores the immense clinical value of placental histopathology. Routine examination of the placenta in all SGA deliveries can provide a definitive etiological diagnosis distinguishing, for example, between MVM secondary to hypertensive disease and villitis from a maternal infection which is critical for counseling on risks in subsequent pregnancies. 40
The overall prevalence of placental abnormalities (47.7%) in this cohort, while substantial, is lower than the rates of over 80% reported in some studies from Ireland and the United States.18,41 This difference may be attributable to variations in population characteristics, definitions of pathology, or sampling protocols. Notably, the prevalence of inflammatory lesions (8.6%) in our study is clinically significant given the endemic burden of maternal infections in Uganda, which are known to cause placental inflammation and are strongly linked to FGR and preterm birth.19,23
The emergence of maternal obesity as a strong independent risk factor (aRR = 1.61) is a critical finding for public health in sub-Saharan Africa. Globally, maternal obesity is known to induce a systemic state of chronic, low-grade inflammation, insulin resistance, and oxidative stress. 26 This adverse metabolic milieu directly impairs placental function by promoting lipotoxicity, endothelial dysfunction, and aberrant inflammatory signalling at the maternal-fetal interface.27,28 Histologically, these processes manifest as lesions of MVM, including syncytial knots, fibrinoid deposition, and infarction, which compromise placental perfusion and nutrient transport. 42 Our finding aligns with the growing recognition of the “double burden” of malnutrition in Uganda, where rising rates of overweight and obesity coexist with persistent undernutrition, posing a complex challenge for maternal and child health services.24,43
The association of higher maternal income with an increased risk of adverse outcomes (aRR = 1.66) is a paradoxical finding that contradicts the vast body of literature where low socioeconomic status is a well-established risk factor for poor perinatal health.44,45 This unexpected result likely reflects complex socioeconomic dynamics specific to our study context. Two primary explanations are plausible. First, this finding may be an indicator of the nutritional transition, where increased economic resources in urbanizing LMIC settings are associated with the adoption of sedentary lifestyles and diets rich in processed, energy-dense foods. 46 In this scenario, higher income acts as a proxy for increased exposure to obesogenic risk factors, which, as this study also demonstrates, is a direct cause of adverse outcomes.
Second, the finding may be influenced by referral bias. The study was conducted in two regional referral hospitals, which are the primary destinations for the most complex obstetric cases from a wide catchment area. Women with higher socioeconomic status may have greater financial resources and social capital, enabling them to overcome significant transport and logistical barriers to reach these tertiary facilities when complications arise. 47 Consequently, this could lead to a concentration of high-risk, complicated pregnancies among higher-income women within this specific hospital-based cohort, creating a spurious association between wealth and poor outcomes.
Placental pathology should be interpreted here as a proximate pathological marker of fetal compromise rather than as an upstream maternal risk factor, and its inclusion in the model was intended for prognostic, not causal, interpretation.
4.2. Strengths and limitations
Strengths of this study include its prospective cohort design, inclusion of two referral hospitals, and standardized placental histopathology performed by a pathologist blinded to neonatal outcomes. However, the hospital-based referral setting may have over-represented complicated pregnancies and NICU admissions, limiting generalizability to lower-level facilities and the community. The study was powered primarily to estimate the incidence of adverse outcomes among SGA births, so regression findings should be interpreted as exploratory/prognostic associations. Although SGA was determined clinically using birth weight and documented gestational age, gestational-age and percentile documentation were not sufficiently standardized to support uniform retrospective recalculation of birth-weight centiles or gestational-age–stratified analyses. Placental lesion subtypes were reviewed histologically, but sparse counts and inconsistent analysable coding limited lesion-specific regression analyses; these were therefore summarized descriptively. As an observational study, causal inference cannot be made.
5. Conclusions
Small-for-gestational-age births in northern Uganda are associated with a very high burden of neonatal morbidity, with nearly half of all SGA infants experiencing a significant adverse outcome. This burden is driven primarily by underlying placental dysfunction, which serves as the final common pathway for maternal insults. Maternal obesity is a key modifiable risk factor that contributes to this placental pathology. These findings strongly support the integration of routine placental examination for all SGA deliveries into clinical practice at referral centers to aid in diagnosis and counselling. Furthermore, they highlight an urgent public health imperative to develop and implement antenatal care strategies that address the emerging epidemic of maternal obesity alongside persistent challenges of maternal infections to improve neonatal survival in Uganda and similar settings.
Footnotes
Acknowledgements
The authors sincerely thank the staff and management of Lira Regional Referral Hospital, Kiryandongo General Hospital, and the Department of Pathology at Mulago National Referral Hospital for their collaboration. Appreciation is also extended to the study participants for their invaluable contribution to this research.
Ethical considerations
Ethical approval for this study was obtained from the Research Ethics Committee of Kampala International University (KIU-REC-2025-849) and clearance was granted by the Uganda National Council for Science and Technology. Administrative permission was also obtained from Lira Regional Referral Hospital and Kiryandongo General Hospital before commencement of the study.
Consent to participate
Written informed consent was obtained from all participants before enrolment into the study. For participants who could not read or write, the information sheet and consent form were explained in their preferred language by a trained research assistant. Sufficient time was provided for participants to ask questions and make an informed decision about participation. Those who agreed to participate signed or provided a thumbprint on the consent form in the presence of an independent literate witness.
Author contributions
Conceptualization and Methodology: Ahmed Ali Fahma, Musa Kasujja, Theoneste Hakizimana. Investigation and Data Curation: Ahmed Ali Fahmo, Marie Pascaline Sabine Ishimwe, Sawda Abdikarim Sheikh Isse, Babikir Abdallah Araba Rajab, Abdifatah Abdullahi Mohamed. Formal Analysis and Interpretation: Musa Kasujja, Theoneste Hakizimana. Writing – Original Draft Preparation: Emmanuel Okurut, Rogers Kajabwangu, Marie Pascaline Sabine Ishimwe and Theodore Nteziyaremye. Writing – Review and Editing: Musa Kasujja, Theoneste Hakizimana, Eltayed Mohamed Ahmed. Supervision: Musa Kasujja, Theoneste Hakizimana. All authors have read and approved the final manuscript.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets generated and analysed during the current study are available from the corresponding author upon reasonable request. Theoneste Hakizimana (