School of Public Health (Public-Health) Collections
Permanent URI for this collection
Browse
Recent Submissions
1 - 5 of 1034
-
ItemPrevalence of viral load non-suppression and associated factors among adult persons living with HIV on antiretroviral therapy at Kigungu Art Clinic, Entebbe, Wakiso District, Uganda(Makerere University, 2026)The human immunodeficiency virus (HIV) epidemic remains a global public health problem despite the increased rollout of antiretroviral therapy (ART). In Uganda, only 75.4% of persons living with HIV (PLHIV) on ART achieve viral load suppression. This figure falls short of the Joint United Nations Programme on HIV/AIDS (UNAIDS) 95-95-95 target, which requires 95% of PLHIV on ART to be virally suppressed. To determine the prevalence of viral load non-suppression and associated factors among adult PLHIV receiving ART at Kigungu Health Center III ART clinic in Entebbe, Wakiso District, Uganda. This cross-sectional study utilized patient clinic records and structured questionnaire interviews. A systematic random sampling technique was employed to select 234 participants for inclusion in this study between June and September 2025. The primary outcome was viral load non-suppression. Data were collected using a structured questionnaire to assess socio-demographics and psychological factors. Drug and clinical factors data were extracted from records, entered into Epidata, and subsequently exported to Stata version 18 for analysis. Prevalence was reported as a percentage, and modified Poisson regression analysis was used to determine associated factors. The study enrolled 234 participants with a median age (Interquartile range [IQR]) of 36 (29, 44) years; 59.2% were female, 56.9% were fisher folks, 47.6% had primary-level education, and 47.8% were Catholics. The prevalence of viral load non-suppression was 19.7% (95% CI:15.0- 25.3). Duration on current ART regimen (aPR 1.02, 95% CI:1.01- 1.03), having a recent CD4 Count <200 cells/µL (aPR 2.21, 95% CI:1.26 - 3.87), and being a current smoker (aPR 2.42, 95% CI: 1.07-5.47) were significantly associated with viral load non-suppression. One in every five persons living with HIV on ART may have viral load non-suppression, posing a significant challenge to achieving the third 95% UNAIDS target. Duration of the current ART regimen, recent low CD4 count, and smoking were significantly associated with viral load non-suppression. There is a need to strengthen existing interventions and design new strategies to improve viral load suppression. The Ministry of Health should ensure continuous access to CD4 testing at all ART clinics, enhance structured long-term follow-up programs, and promote smoking cessation among PLHIV on ART.
-
ItemEnhancing outbreak surveillance through integration of natural language processing in Uganda’s electronic integrated disease surveillance and response system(Makerere University, 2026)Early detection of diseases or infections is essential to prevent infectious diseases from escalating into large outbreaks. In Uganda, the Electronic Integrated Disease Surveillance and Response (eIDSR) system enables community-level reporting of suspected cases via SMS. However, manual processing of these unstructured messages often delays outbreak detection and response, particularly during high-volume reporting periods. The manual processing of incoming SMS messages within the eIDSR system creates a bottleneck that hinders timely outbreak detection and response. This delay has the potential to increase morbidity and mortality, especially in resource-limited settings. This study aimed to integrate Natural Language Processing (NLP) to automate the extraction of key information, such as disease type, location, and symptoms, from SMS alerts submitted to the eIDSR system. It also sought to understand the contextual factors that influenced model accuracy and performance. A retrospective design was employed using historical SMS data submitted to the eIDSR system in 2024. A Bidirectional Encoder Representations from Transformers (BERT)-uncased model was fine-tuned on a manually annotated dataset to support named entity recognition. The model was evaluated using precision, recall, F1-score, and processing speed, and its performance was compared with manual extraction. McNemar’s test was used to assess the statistical significance of differences between the two methods. The model achieved an F1-score of 92.6%, with recall of 94.2% and precision of 91.1%, processing approximately 48 messages per second. It extracted high-value entities such as disease, age, gender, and location, with near-perfect accuracy. Errors were concentrated around symptom span boundaries and ambiguous entries. Interviews confirmed the value of automation for reducing analyst workload and outlined key limitations of the current manual workflow, including handling of ambiguous or duplicate messages. This study demonstrated the feasibility of applying NLP to automate SMS-based disease surveillance within Uganda’s eIDSR system. Although human review remains necessary for edge cases, the model showed strong potential to accelerate processing, eliminate backlog, and support timely response under frameworks like 7-1-7. With targeted improvements especially in symptom handling and multilingual input. The model would be suitable for pilot integration under a human-in-the-loop deployment model.
-
ItemPrevalence and determinants of unmet blood needs among patients at Jinja Regional Referral Hospital in Eastern Uganda(Makerere University, 2025)Access to safe blood is critical for healthcare delivery, especially in resourcelimited settings. Uganda requires 450,000 units annually but consistently falls short by over 30%. In the Eastern Region, which has the highest maternal mortality ratio nationally, 34% of maternal deaths are linked to a lack of blood. Despite shortages, inappropriate use and wastage occur, and there is limited data quantifying unmet blood needs and their determinants. The study determined the prevalence of unmet blood needs and their determinants among patients at Jinja Regional Referral Hospital in Eastern Uganda. Methods: A convergent mixed-methods design was employed at Jinja Regional Referral Hospital, Uganda, in July 2025. Quantitative data were obtained via a census of 987 blood transfusion orders from the hospital’s electronic A-LIS system to estimate the prevalence of unmet blood needs. Qualitative data were collected through 25 key informant interviews and 17 in-depth patient interviews, purposively sampled to explore systemic, institutional, and socio-cultural factors influencing unmet blood needs. Quantitative analysis used descriptive statistics while qualitative data underwent hybrid thematic analysis, with findings triangulated across sources to ensure validity. Results: A total of 987 blood transfusion orders at Jinja RRH were analyzed; the majority, 66.7% were for females, while 33.3% were for males, and the mean patient age was 32 years (SD 19). Most orders were for blood group O+ (42.2%). Overall, 68.3% of orders were unmet, highest among females (69.6%), patients aged 20–30 years (39.9%), and those with blood group O- (83%). Qualitative findings revealed institutional factors influencing unmet blood needs, such as high demand, stock-outs, off-site screening for TTIs, inadequate storage capacity, staffing shortages, and community barriers, including low donations, mistrust, and religious beliefs. Conclusion: The high proportion of unmet blood transfusion orders (68.3%) at Jinja RRH, especially for blood group O-, warrants action and solutions to target both institutional and community factors. Inefficiencies in the blood supply chain and low voluntary donation hinder timely access. Targeted blood donation drivers should be considered for high-demand blood groups. Strengthening supply chain management and community mobilization is critical to ensuring adequate and safe blood availability
-
ItemPredictors of particulate matter 2.5 and carbon monoxide levels in government primary schools within Kampala, Uganda(Makerere University, 2026)Exposure to fine particulate matter (PM₂.₅) and carbon monoxide (CO) poses serious health risks, particularly for children with developing respiratory systems. In Kampala, Uganda, schools are vulnerable to emissions from traffic, industries, biomass fuel use, and open waste burning, yet limited research has examined pollutant levels and their predictors in government primary schools. This study assessed the levels of PM₂.₅ and CO in government primary schools in Kampala and examined the predictors of their concentrations. A convergent parallel mixed-methods design was employed, incorporating quantitative and qualitative data from 39 government primary schools across three divisions. Air quality was measured in five classrooms, one playground, and one kitchen in each school using calibrated portable monitors(Temtop M2000C for PM₂.₅; AS8700A for CO) at three time points daily. Structured questionnaires were administered to 74 school staff to capture school-level characteristics and potential predictors. Quantitative data were complemented by focus group discussions with 23 pupils and detailed field observations. The data were analysed in Jamovi (v2.6.44) using descriptive statistics, t-tests, ANOVA, and regression models to identify significant associations, while qualitative data were analysed thematically. Results: PM₂.₅ concentrations surpassed the WHO 24-hour guideline reference value (15 μg/m³) across all measured school environments (p < .001) with mean differences ranging from 22.5 µg/m³ to 221.7 µg/m³. Morning kitchen CO levels surpassed the WHO guideline reference value (4 ppm) by 11.6 ppm (p < .001). Descriptive assessment against the 2024 Uganda National Air Quality 24-hour guidelines ( 35 μg/m³ for PM₂.₅; 6.11 ppm for CO) similarly showed exceedances for PM₂.₅ across all measured school environments and Morning CO levels in kitchens, respectively. Field observations revealed that biomass fuel use in school kitchens was nearly universal (97.4%), with firewood being the primary fuel (92.3%). Chalk dust was present in all classrooms (100%), and dust from unpaved playgrounds was noted in 76.9% of schools. Structured questionnaires and FGDs highlighted several perceived risk factors, including dusty playgrounds, proximity to busy roads, waste burning within or near school premises, and heavy reliance on biomass fuels in school kitchens. Qualitative data highlighted health concerns, limited resources, and challenges with policy implementation, with a major lack of funding and infrastructure cited as barriers to effective interventions. Linear mixed-effects regression analysis identified time of day as the strongest predictor of both pollutants, with morning concentrations significantly higher than afternoon levels (PM₂.₅: β = 0.586, p < .001; CO: β = 0.742, p < .001). Measurement location explained 25.1% of PM₂.₅ variance and 64.2% of CO variance, with kitchens consistently showing the highest exposures. PM₂.₅ and CO concentrations in Kampala government primary schools surpassed safe thresholds, with several modifiable factors under the time of day as a significant predictor of their levels. Strengthening interventions such as improved ventilation, cleaner cooking fuels, and traffic management around schools is essential to protect children’s health.
-
ItemPrevalence and factors associated with uptake of voluntary medical male circumcision services among men aged 25-39 years in Rakai, Uganda(Makerere University, 2025)Despite significant advances, the global challenge of controlling HIV/AIDS remains, especially in sub-Saharan Africa, which is home to two-thirds of those living with HIV. Uganda’s HIV prevalence stands at 5.8%. To curb transmission, the WHO recommends combination prevention strategies, including voluntary medical male circumcision (VMMC), a cost-effective approach. Engaging men aged 25 and above, who play a significant role in HIV transmission, has proven challenging in Uganda. Currently, 57.5% of Ugandan men aged 15–49 have taken up VMMC, with most uptake among adolescents. Cultural norms, misconceptions, and perceived risks hinder uptake among older men, yet few studies have explored these barriers in Uganda. Objective: This study sought to determine factors influencing VMMC uptake and explore motivators and barriers for men aged 25–39 in Rakai district, aiming to inform targeted HIV prevention efforts. Methods: Using a concurrent mixed-methods design, the study analyzed secondary quantitative data from 2,317 men aged 25–39 from the Rakai Community Cohort survey from Decemeber 2020 to March 2023. The modified Poisson model in STATA v14 was employed to determine factors associated with VMMC uptake, while qualitative data from 21 in-depth and key informant interviews was analyzed in atlas.ti to provide insights into motivators and barriers to uptake. Results: Voluntary medical male circumcision uptake among men aged 25–39 was 64.1%, lower than the 79.2% among younger men aged 15–24. Factors linked to uptake included being older than 35 years, having 2 or mor sexual partners in 12 months, never using condoms, and never having tested for HIV. Key motivators were STI prevention and improved hygiene. Barriers included work-related time constraints, pain, delayed healing, and limited communication targeting older men. Conclusion: The uptake of VMMC among men aged 25–39 in Rakai remains suboptimal, despite their key role in HIV epidemic control. Implementing communication strategies such as one-to one engagements preferred by older men and modifying the VMMC service to address concerns of loss of working time may increase VMMC uptake among older men to acceptable levels.