Mbarara University of Science and Technology Institutional Repository (MUST-IR)

MUST-IR preserves research output from the MUST Community

Communities in MUST-IR

Select a community to browse its collections.

Now showing 1 - 10 of 10

Annual Research Dissemination Conference 3
Central Administration 124
This community will contain collections from the general Administrative staff
Faculty of Applied Sciences and Technology 54
Faculty of Business and Management Sciences 107
Faculty of Computing and Informatics 137
Faculty of Interdisciplinary Studies 210
Faculty of Medicine 2254
Faculty of Science 538
Institute of Tropical Forest Conservation 126
Maternal Newborn and Child Health Institute 18

Recent Submissions

Participation of livestock farmers and community animal health workers in one health intervention to control zoonotic diseases in Rwanda: a qualitative study
(BMC Infectious Diseases, 2026) Shema Hugor; Jonathan Izudi; Anselme Shyaka
Background: In zoonotic disease hotspots, engaging local communities in zoonotic disease monitoring and reporting is essential for effective prevention and early outbreak detection. In Rwanda, the Eastern Province has experienced nearly 90% of all zoonotic disease outbreaks. This qualitative study explored the participation of livestock farmers and community animal health workers (CAHWs) in One Health interventions aimed at controlling zoonotic disease outbreaks in Nyagatare district, Eastern Province, Rwanda. Method: Among livestock farmers and CAHWs, we collected qualitative data through in-depth interviews and focused group discussions. We held key informant interviews with district-level officials, namely district veterinary, district health, and district environmental health officers, to validate the data from livestock farmers and CAHWs. The data were transcribed verbatim and verified by replaying the audio recordings. Content analysis was performed. Results: Five major themes emerged as areas of participation: (i) capacity building focused on livestock management and livestock disease identification and prevention; (ii) community-level health education encompassing zoonotic disease risk communication and awareness raising; (iii) interruption of zoonotic disease transmission that comprised infection prevention and control, separation of sick and healthy livestock, quarantine, and zero grazing, adherence to guidelines, and the testing, treatment, spraying, and vaccination of livestock; (iv) collaboration with multiple sectors via training on zoonotic diseases, including its treatment, investigation, and prevention; and 5) zoonotic disease surveillance that included livestock monitoring and inspection for zoonotic diseases and the notification and reporting of zoonotic diseases to relevant authorities. The collaboration between the environmental health sector and the animal and human sectors was weak, while vaccination targeted mainly cattle but not small ruminants. Conclusions: Livestock farmers and CAHWs participate in several One Health interventions aimed at controlling zoonotic diseases, particularly in vaccination campaigns for cattle, community sensitization, and basic animal husbandry practices. However, collaboration between the environmental health sector and the animal and human sectors was weak, vaccination efforts largely excluded small ruminants, and some farmers relied on local herbs for treatment due to limited access to professional veterinary services. Efforts to sustain the major areas of participation are needed, while weaker areas need strengthening.
The impact of selection criteria on the properties of green valley galaxies
(Astronomy & Astrophysics, 2026) Beatrice Nyiransengiyumva; Mirjana Povic; Pheneas Nkundabakura; Tom Mutabazi; Antoine Mahoro
Context: The bi-modality in the distribution of galaxies usually obtained from colour-colour or colour-stellar mass (absolute magnitude) diagrams has been studied to determine the difference between galaxies in the blue cloud and in the red sequence, as well as to define the green valley region. As a transition region, green valley galaxies can offer clues about the morphological transformation of galaxies from late-type to early-type. Therefore, the selection of green valley samples is of fundamental importance. Aims: In this work, for the first time, we evaluate the selection effects of the most frequently applied green valley selection criteria. The aim is to understand how these criteria affect the identification of green valley galaxies, their properties, and their impact on galaxy evolution studies. Methods: Using the SDSS optical and GALEX ultraviolet data at redshift z < 0.1, we selected the eight most commonly used criteria based on colours (without and with Gaussian fittings), specific star formation rate, and star formation rate versus stellar mass. We then studied the properties of the green valley galaxies (e.g. their stellar mass, star formation rate, specific star formation rate, intrinsic brightness, and morphological and spectroscopic types) for each selection criterion. Results: We found that when using different criteria, we selected different types of galaxies. UV-optical colour-based criteria tend to select more massive galaxies, with lower star formation rates and a higher fractions of composite and elliptical galaxies than when using pure optical colours. Our results also show that the colour-based criteria are the most sensitive to galaxy properties, rapidly changing the selection of green valley galaxies. Conclusions: Whenever possible, we suggest avoiding the green valley colour-based selection and using other methods or a combination of several, such as the star formation rate versus stellar mass or specific star formation rate.
Longitudinal Trends in Physical Activity Among Older Adults With and Without HIV in Uganda
(Journal of aging and health, 2026) Aneeka Ratnayake; Yao Tong; Zahra Reynolds; Steffany Chamut; Lien T. Quach; Phoebe Mbabazi; Shruti Sagar; Samuel Maling; Crystal M. North; Eliza Passell; Moka Yoo-Jeong; Alexander C. Tsai; Robert Paul; Christine S. Ritchie; Janet Seeley; Susanne S. Hoeppner; Flavia Atwiine; Edna Tindimwebwa; Samson Okello; Noeline Nakasujja; Deanna Saylor; Meredith L. Greene; Stephen Asiimwe; Jeremy A. Tanner; Mark J. Siedner; Brianne Olivieri-Mui
Introduction: Physical Activity (PA) and its links to frailty, quality of life (QoL), and other comorbidities in older Ugandans living with HIV remain under-explored. Methods: We analyzed data from three annual assessments of older people living with HIV (PLWH) and age- and sex-similar people not living with HIV (PnLWH). We fitted linear generalized estimating equations (GEE) regression models to estimate the correlates of PA, including demographics, frailty, QoL, HIV, and other comorbidities. Results: We enrolled 297 PLWH and 302 PnLWH. Older age (b = 157.34, 95% CI [-222.84, 91.83]), living with HIV (b = 979.88 [95% CI: 1878.48, 81.28]), frailty (b = 3011.14 [95% CI: 4665.84, 1356.45]), and comorbidities (b = 2501.75 [95% CI: 3357.44, 1646.07]) were associated with lower overall PA. Higher general QoL (b = 89.96 [95% CI: 40.99, 138.94]) was associated with higher PA. Conclusion: PA interventions may support wellbeing of older people in the region, and tailored interventions should be explored.
Extreme weather effects on health services and communities in low and lower-middle income countries: a thematic systematic review
(Transactions of The Royal Society of Tropical Medicine and Hygiene, 2026) Julii Brainard; Yovitha Sedekia; Natalia R. Jones; Michael Matte; Patrick Sunday; Deborah Watson-Jones; Daniel Dennis Mapemba; Severin A. Kabakama; Edgar Mugema Mulogo; Moses Ntaro; Tran Thi Tuyet-Hanh; Paul R. Hunter; Jo-Anne Geere
Most previous research about the dangers of extreme weather events was applicable to populations in high income countries. Data summarising harms related to extreme weather events in low-income settings are lacking. A systematic review thematically summarising evidence about weather event-linked harms and responses in low- and lower-middle-income countries was conducted. Peer-reviewed and grey literature was systematically searched and selected. Data were extracted about harms, responses and outcomes relevant to six WHO building blocks of healthcare systems. Framework analysis was used to identify predominant themes related to harms, responses and the WHO building blocks. In total, 183 reports were included. Flooding and high winds were the most common types of extreme weather events documented. The main community experience themes identified were the displacement of populations and disruption. The main themes identified for health service delivery were vulnerability, disruption and resilience. Documented examples of resilience or recovery were far fewer for all six WHO healthcare system building blocks than descriptions of vulnerability and disruption. Extreme weather events can be highly disruptive and harmful to healthcare systems and communities in LMIC settings that are often already highly vulnerable
Weighted average algorithm adjusted a novel (1+FOPI)-FOPI-TID controller structure for AGC with integration of non-linearities and cyber-attack
(Scientific Reports, 2026) Moses Awal; Michael Robson Atim; Jimmy Nabende Wanzala; Johnes Obungoloch; Mohamed Barakat
The integration of diverse energy sources and the advent of smart grids have intensified the challenges in load frequency management (LFM). Modern power systems are increasingly vulnerable to inherent nonlinearities, such as generation rate constraints, governor dead bands, boiler dynamics, and communication delays, as well as sophisticated cyber-attacks, which collectively threaten frequency stability and tie-line power balance. To address these challenges, this study proposes a novel cascade controller, designated as (1+FOPI)-FOPI-TID, for robust automatic generation control in hybrid two-area power systems. The controller uniquely combines fractional-order (FO) dynamics with a tilt-integral-derivative stage and is optimized using a green metaheuristic, the weighted average algorithm (WAA). The WAA effectively balances exploration and exploitation to achieve superior parameter tuning. The proposed control architecture processes both area control error (ACE) and frequency deviation (ΔF) signals through dedicated stages, enabling enhanced disturbance rejection and transient response. The system model incorporates a comprehensive set of nonlinearities and evaluates resilience against resonance-based cyber-attacks. Comprehensive simulation studies under both AC and HVDC tie-line configurations demonstrate that the WAA-optimized (1+FOPI)- FOPI-TID controller significantly outperforms existing schemes, including PD-PI, PIFOD-(1+PI), and PIDF(1+FOD). Key performance metrics show a 45.3% reduction in the integral of time weighted absolute error (ITAE) and improvements in settling times of 47.7% for ΔF₁ and 32.8% for ΔF₂. Sensitivity analysis confirms robustness under±25% parameter variations and random load perturbations. During cyber-attacks, the controller maintains the lowest Rate of Change of Frequency (RoCoF), underscoring its dual capability in stabilizing grid dynamics and mitigating cyber-physical threats. These results validate the controller’s potential to enhance operational resilience and reliability in future smart grids.
Fecal carriage and molecular characterization of ESBL-producing Enterobacteriaceae among farmers in Mid-Western Uganda
(Scientific Reports, 2026) Galimaka Wilson; Kabera Micheal; Abaasa Catherine; Nalumaga Pauline Petra; Fredrickson B. Wasswa; Kassaza Kennedy; Ampaire Lucus; Ampaire Lucus
Extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae pose a growing threat to public health due to resistance to β-lactam antibiotics. This study aimed to determine the prevalence, antimicrobial resistance patterns, and genetic characterization of ESBL-producing Enterobacteriaceae from stool samples of farmers in Kibimba Parish, Kabarole District, Mid-Western Uganda. A crosssectional study among 250 farmers involved stool culture, ESBL detection via the double-disc synergy test, PCR for blaCTX-M, blaTEM, and blaSHV genes, and antibiotic susceptibility testing. Data were analyzed using STATA v14.2. The prevalence of ESBL-producing Enterobacteriaceae was 36.4% (91/250) among farmers and 34.29 (107/312) (95%CI: 29.22–34.76) among isolated organisms. Escherichia coli was the predominant ESBL-producing isolate at 70.4% (76/107). High resistance was observed to piperacillin (94.0%) and ampicillin (81.4%), while imipenem (96.5%) and chloramphenicol (94.0%) showed the highest susceptibility. Multidrug resistance (MDR) was noted in 49.4% of all isolates and 33.7% of ESBL producers. Among 107 confirmed ESBL producers, 80.4% harbored at least one ESBL gene; blaCTX-M (48.7%), blaTEM (34.9%) and blaSHV (16.4%). ESBL carriage was significantly associated with use of shallow well water (p<0.009), goat farming (p<0.036), and chronic illness (p<0.012). Farmers showed high fecal carriage of multidrug-resistant ESBL-producing Enterobacteriaceae, linked to environmental and health-related risk factors.
Inpatient initiation of tuberculosis preventive therapy with 1 month of isoniazid and rifapentine for adults with advanced HIV disease and cryptococcal meningitis (IMPROVE): a non-inferiority, randomised controlled trial
(The Lancet HIV, 2026) Jayne Ellis; Gila Hale; Laura J Nsangi; Abduljewad Wele; Enos Kigozi; Jane Gakuru; Enock Kagimu; Timothy Mugabi; Suzan Namombwe; Sarah Kimuda; Faizo Ssekindi; Jane F Ndyetukira; Alisat Sadiq; Asmus Tukundane; Wilber Bakka; Thomas S Harrison; Emmanuel Mande; Conrad Muzoora; David A J Moore; David B Meya; Katherine Fielding; David R Boulware; Joseph N Jarvis
Background: Tuberculosis preventive therapy coverage for people with advanced HIV disease (AHD) is poor. Innovative delivery strategies to increase tuberculosis preventive therapy uptake are needed; we sought to evaluate the safety and feasibility of two strategies for ultra-short course tuberculosis preventive therapy with 1 month of daily rifapentine plus isoniazid (1HP). Methods: In this phase-3, open-label, non-inferiority, randomised controlled strategy trial (ISRCTN 18437550), we recruited consecutive adults (aged ≥18 years) admitted to hospital with AHD receiving treatment for cryptococcal meningitis who were screened for active tuberculosis during their hospitalisation from three tertiary referral hospitals in Uganda (Mulago National Specialised Hospital, Kiruddu National Referral Hospital in Kampala, and Mbarara Regional Referral Hospital). Adults without evidence of tuberculosis disease and meeting all eligibility criteria were approached for consent and inclusion. Patients were excluded if they had evidence of active hepatitis B infection, abnormal liver function tests, had known chronic liver disease, were jaundiced, were pregnant or breastfeeding, or presented with a clinical syndrome which, in the opinion of the attending clinician, put the patient at significant risk if they were to participate in the trial. After providing informed consent, we randomly assigned participants (1:1) to inpatient initiation of 1HP before hospital discharge or outpatient initiation at 6 weeks after time of cryptococcal meningitis diagnosis. 1HP was standardised across treatment groups, a 28-day course of 600 mg rifapentine plus 300 mg isoniazid daily with adjunctive pyridoxine (25 mg per day). The 1HP regimen was not dose adjusted on the basis of weight. The primary endpoint was tuberculosis disease-free survival and 1HP treatment completion at 18 weeks, powered for a 15% non-inferiority margin; analysis was by intention to treat. Findings: From Jan 24, 2022, to Nov 13, 2024, 419 adults were screened after 210 were found ineligible and four died before random allocation, 205 were randomly allocated (171 in Kampala and 34 in Mbarara, Uganda): 103 to the inpatient group and 102 to the outpatient group. 119 participants (58%) were male and 86 (42%) were female. In the primary adjusted intention-to-treat analysis, 72 participants in the inpatient 1HP group (70%) had tuberculosis disease-free survival and 1HP treatment completion at 18 weeks compared with 63 (62%) in the outpatient 1HP group (adjusted risk difference 7·1%, 90% CI –3·8 to 17·9) confirming non-inferiority. Treatment completion was achieved in 78 (76%) of 103 in the inpatient 1HP group compared to 67 (66%) of 102 in the outpatient 1HP group (site-adjusted risk difference 9·7%, 95% CI –2·4 to 21·8). 170 grade 3 or 4 adverse events occurred in 99 (48%) of 205 participants. Among participants who had taken at least one dose of 1HP the frequency of adverse events across trial groups was similar apart from grade 4 anaemia, which occurred in a higher proportion of participants in the outpatient group (9% vs 2%, p=0·045). Interpretation: 1HP initiation before hospital discharge was non-inferior to outpatient initiation among adults with AHD and cryptococcosis. These data suggest that following exclusion of active tuberculosis disease, inpatient 1HP initiation is feasible and comparably safe compared with outpatient initiation.
In Silico Identification of Antiviral Peptides as Potential Leads Against Sudan Ebolavirus VP-40
(BioMed Research International, 2026) Boniface Omara; Kenedy Kiyimba; Fatoumata G. Fofana; Oudou Diabaté; Walter Odur; Daudi Jjingo; Jacob Stanley Iramiot; Peace Draleru; Joan Achia; Muhammad Shafiq; Zaheer Ul-Haq; Hedmon Okella; Steven Odongo
The continued reemergence of Ebola virus epidemics remains a global health concern, largely due to limited therapeutic interventions. This study is aimed at identifying and characterizing antiviral peptides as potential lead candidates against the Sudan Ebola virus. We retrieved antiviral peptides from the AVPdb and designed novel peptides from them using support vector machine, RF, and discriminant analysis algorithms. The toxicity and allergenicity predictions were performed using ToxinPred, ADMETLab 3.0, Allertop, and AllergenFP web servers, respectively. The 3D structures of selected peptides were modeled using PEP-FOLD and I-TASSER and validated using ProSA and PROCHECK web servers.The best peptide models were docked against the Sudan Ebola virus VP-40 protein using HDOCK and ClusPro. Molecular dynamics (MD) simulations were then carried out in GROMACS 2024.2. Out of 170 designed motifs, 30 exhibited antiviral potential with antiviral scores ranging from 0.506 to 1.000. Among the predicted antiviral peptides, five demonstrated favorable stabilities, nontoxicity, and non-allergenic properties. PEP-FOLD produced more stable peptide structures than I-TASSER, with over 84.6% of their amino acids in the most favorable region. Binding energies ranged from −252.39 to −145.83 kcal/mol (HDOCK) and from −887.7 to −538.7 units (ClusPro). The MD simulations confirmed high stability, with motif A10_M showing the strongest binding and structural compactness. Five peptides show strong potential as therapeutic leads against Sudan Ebola virus; however, further experimental validation is recommended.
Beyond overconfidence: Embedding curiosity and humility for ethical medical AI
(PLOS Digital Health, 2026) Sebastián Andrés Cajas Ordóñez; Rowell Castro; Leo Anthony Cel; Roben Delos Reyes; Justin Engelmann; Ari Ercole; Almog Hilel; Mahima Kalla; Leo Kinyera; Maximin Lange; Torleif Markussen Lunde; Mackenzie J. Meni; Anna E. Premo; Jana Sedlakova
Contemporary medical AI systems exhibit a critical vulnerability: they deliver confident predictions without mechanisms to express uncertainty or acknowledge limitations, leading to dangerous overreliance in clinical settings. This paper introduces the BODHI (Bridging, Open, Discerning, Humble, Inquiring) framework, a dual-reflective architecture grounded in two essential epistemic virtues: curiosity and humility, as foundational design principles for healthcare AI. Curiosity drives systems to actively explore diagnostic uncertainty, seek additional information when faced with ambiguous presentations, and recognize when training distributions fail to match clinical reality. Humility provides complementary restraint, enabling uncertainty quantification, boundary recognition, and appropriate deference to human expertise. We demonstrate how these virtues function synergistically in a dynamic feedback loop, preventing both reckless exploration and excessive caution while supporting collaborative clinical decision-making. Drawing from psychological theories of curiosity and cross-species evidence of epistemic humility, we argue that these capacities represent fundamental biological design principles essential for systems operating in high-stakes, uncertain environments. The BODHI framework addresses systemic failures in medical AI deployment, from biased training data to institutional workflow pressures, by embedding uncertainty awareness and collaborative restraint into foundational system architecture. Key implementation features include calibrated confidence measures, out-of-distribution detection, curiosity-driven escalation protocols, and transparency mechanisms that adapt to clinical context. Rather than pursuing algorithmic perfection through pure optimization, we advocate for human-AI partner ships that enhance clinical reasoning through mutual accountability and calibrated trust. This approach represents a paradigm shift from overconfident automation toward collaborative systems that embody the wisdom to pause, reflect, and defer when appropriate.
Assessment of prescribing practices for respiratory tract infections in public health facilities, Jinja City, Uganda, June 2022–May 2023
(Discover Public Health, 2026) Zablon K. Igirikwayo; Humphreys Mukaga; Richard Migisha
Background: Rational drug use ensures that medications are prescribed based on clinical needs, appropriate dosages, and for the right duration, minimizing healthcare costs. Irrational antibiotic use is a key driver of antibiotic resistance (ABR) emergence. We evaluated the conformance of drug prescriptions to the World Health Organization (WHO) prescribing indicators among patients with respiratory tract infections (RTIs) attending public health facilities in Jinja City, Uganda. Methods: A retrospective observational study was conducted across 11 public health facilities in Jinja City, Eastern Uganda, from June 1, 2022, to May 31, 2023. Patient records of those diagnosed with RTIs were selected using systematic random sampling. Several prescribing indicators were assessed, including the number of drugs prescribed per patient encounter, the percentage of encounters with antibiotics, the percentage of drugs from the essential drugs list, the percentage of drugs prescribed by generic name, and the percentage of encounters with injectable medications. Data were analyzed for compliance with WHO prescribing standards. Results: Three indicators met or were close to WHO targets: the average number of drugs prescribed per patient (2.7, WHO target<3), the percentage of encounters with injections prescribed (3.0%, WHO target≤10%), and the percentage of drugs from the essential drugs list (93.4%, WHO target 100%). However, two indicators did not meet WHO recommendations: the percentage of encounters with antibiotics prescribed (79.8%, WHO target<30%) and the percentage of drugs prescribed by generic name (79.5%, WHO target 100%), suggesting irrational prescribing practices. Among patients prescribed antibiotics, most received only one antibiotic (89.1%). Conclusion: This study revealed gaps in adherence to WHO prescribing standards, particularly regarding excessive antibiotic prescriptions for RTIs. These irrational prescribing practices may contribute to the emergence of ABR. There is a need for targeted interventions to promote rational drug use and improve prescribing practices across health facilities in the region.
Factors linked to extramarital sex and its relationship with HIV infection: a cross sectional analytical study in Southwestern Uganda
(AIDS Research and Therapy, 2026) Prisca Asiimwe; Grace Nambozi; Ronald Kamoga
Background: The rising prevalence of extramarital sexual networking has led to increasing susceptibility to HIV infection among married couples. This study aimed to determine the prevalence of extramarital sex among married individuals in Mbarara Regional Referral Hospital, the factors linked to it and its relationship with HIV. Methods: Data were collected using an interviewer-administered questionnaire. Participants were clients seeking HIV testing at a tertiary hospital in southwestern Uganda between June and September 2025. All clients received standard pre- and post-test counselling. Data was analyzed using Stata version 17. Results: The study included a sample of 384 participants, with an almost equal distribution across the genders (50.3% female; 49.7% male). The prevalence of extramarital sex was 58.1% (95% CI: 0.51–0.65) among men and 8.2% (95% CI: 0.05–0.13) in women. Among those who reported extramarital sex, the prevalence of HIV was 10.5% (95% CI: 0.07– 0.16) and 5.1% (95% CI: 0.03–0.13) for men and women respectively. An equal number (8) of men and women who denied involvement in extramarital sex were also diagnosed with HIV. The results suggested no relationship between extramarital sex and HIV (OR 0.9, 95% CI: 0.38–1.90, p=0.7). However, women were found to be at an increased risk of acquiring HIV (RR=2.02, 95% CI: 0.91–4.47) compared to men. Multivariate logistic regression identified the following factors to be linked to extramarital sex: being male (aOR=16.4, 95% CI: 9.13–29.41, p<0.001), monogamous marriage status (aOR=3.2, 95% CI: 1.51−6.93, p=0.002), marital duration exceeding 19 years (aOR=3.8, 95% CI: 1.92−7.56, p<0.001), being Muslim (aOR=2.4, 95% CI: 1.04−5.49, p=0.04), alcohol consumption (aOR=3.5, 95% CI: 2.14−5.58, p<0.001), sexual dissatisfaction among females (aOR=11.1, 95% CI: 3.82–15.14, p<0.001) and sharing workplace (aOR=7.6, 95% CI: 0.96–61.01, p=0.05). Conversely, factors negatively linked to extramarital sex in this study included identifying as female (aOR=0.7, 95% 0.03–0.14, p<0.001), and having no formal education (aOR=0.3, 95% CI: 0.09−0.63, p=0.003). Conclusion: The study observed a high rate of extramarital sex in the study setting. This was a baseline study. Therefore, subsequent research is needed to identify the underlying reasons for seeking extramarital affairs in Southwestern Uganda.
Factors influencing antibiotic prescription for respiratory tract infections among prescribers in Jinja City, Uganda: a qualitative study
(BMC Pulmonary Medicine, 2026) Zablon K. Igirikwayo; Edison Byamugisha; Jerome Kabakyenga
Background: Inappropriate antibiotic prescription for respiratory tract infections (RTIs) is a major driver of antimicrobial resistance (AMR). In Uganda, 40–80% of RTI patients receive antibiotics (ABs) despite most of them being viral and self-limiting. Few studies have explored the multifaceted factors influencing prescribing behavior among Ugandan healthcare providers. This study aimed to identify factors influencing antibiotic prescription (ABP) for RTIs among prescribers in Jinja City, Uganda. Methods: We conducted an exploratory qualitative study using face-to-face in-depth interviews with 16 prescribers (10 medical clinical officers, 5 nurses, 1 medical officer) from 10 purposively selected public health facilities in Jinja City during June-July 2023. Interview topics included knowledge of RTI etiology, antibiotic (AB) prescribing practices, availability of standard treatment guidelines (STGs) and diagnostics, and continuous education on AB use. Audio recorded responses were transcribed verbatim, coded, and analyzed thematically using the Social Ecological Model to identify multilevel factors influencing prescribing decisions. Results: The factors influencing ABP for RTIs were categorized into individual, interpersonal, institutional, community and ministry of health (MOH) and government policy factors. Individual-level factors included prescriber related factors and state of the patient considerations. Interpersonal factors involved prescriber-patient relationship and interactions with drug sales representatives. Institutional factors included initiatives to promote rational AB use by the facility’s management. Community-level factors included AB access and socioeconomic environment. Ministry of health and government policy factors included healthcare delivery infrastructure, educational and training programs, STGs, enforcement of regulations for prescription and dispensing. Conclusions: Multiple interrelated factors at individual, interpersonal, institutional, community, and policy levels contribute to inappropriate AB prescribing for RTIs in Jinja City. Addressing this challenge requires multipronged interventions. These findings provide actionable evidence for developing context-specific interventions to promote rational AB use and combat ABR in Uganda and similar low-resource settings.
Prevalence and associated factors of intradialytic hypotension among patients with kidney failure on maintenance hemodialysis at Kiruddu National Referral Hospital, Uganda
(BMC nephrology, 2026) Dauglas Nkoyooyo; Daniel Kiggundu; Derrick Asiimwe; Conrad Lubwama; Martin Ssentongo; Didas Mwebesa; Benard Owere; Rosemary Namayanja; Ronald Ouma Omolo; Collin David Agaba; Richard Migisha
Background: Intradialytic hypotension (IDH), a frequent hemodialysis complication, is associated with increased morbidity, reduced quality of life, and increased mortality in patients with kidney failure. However, data on the burden and associated factors of IDH in sub-Saharan African settings, including Uganda are limited. This study determined the prevalence and associated factors of IDH among patients with kidney failure attending the dialysis unit of Kiruddu National Referral Hospital (KNRH), Uganda. Methods: A cross-sectional study enrolled adult patients (aged≥18 years) with kidney failure undergoing maintenance hemodialysis through arteriovenous fistula or a central venous access catheter at KNRH from November 2024 to January 2025. Each participant was recruited and observed only once, during a single hemodialysis session. Blood pressure (BP) was measured 5 min pre-, intra- (every 30 min), and 5 min post-dialysis. IDH was defined as per European Best Practice Guidelines (EBPG: systolic blood pressure [SBP] decrease>20 mmHg or mean arterial pressure [MAP]>10 mmHg with symptoms requiring intervention) or nadir criteria (SBP<90 mmHg). Sociodemographic, clinical, and dialysis-related data were collected using interviewer-administered questionnaire and from medical records review. We used multivariable logistic regression to identify associated factors. Results: We enrolled 186 participants with a median age of 46.5 (inter-quartile range [IQR]=34–58) years; 56.9% were male. Overall, 25 developed IDH for a prevalence of 13.4% (95% CI: 9–19%). EBPG and nadir definitions identified 18 (9.67%) and 19 (10.22%) cases of IDH, respectively. Heart failure (aOR=6.41, 95% CI: 1.29–31.77, p=0.023), diabetes mellitus (aOR=4.85, 95% CI: 1.15–20.40, p=0.031), and low hemoglobin (<6 g/dL) (aOR=6.35, 95% CI: 1.91–21.15, p=0.003) were significantly associated with IDH Conclusion: Intradialytic hypotension was relatively common at KNRH, affecting one in ten patients with kidney failure, and was significantly associated with underlying comorbidities, including heart failure, diabetes, and severe anemia. Early identification and management of these risk factors—through optimized volume status, improved glycemic control, and correction of severe anemia—may mitigate IDH episodes and enhance outcomes for patients with kidney failure on hemodialysis. Future longitudinal studies could assess clinical and prognostic implications of IDH in this setting.
Predicting multi-factor authentication uptake using machine learning and the UTAUT framework
(Academia AI and Applications, 2026) Ronald Kato; Aggrey Obbo; Richard Kimera
This study investigates a machine learning-driven framework for predicting multi-factor authentication (MFA) adoption in Uganda’s financial services ecosystem, a context increasingly exposed to cybersecurity risks as digital finance expands. This research aims to (i) identify key behavioral, technological and contextual determinants influencing MFA uptake, (ii) develop and validate an interpretable predictive model aligned with the Unified Theory of Acceptance and Use of Technology (UTAUT) and (iii) compare multiple classification algorithms, including classical ensembles and custom neural architectures, to establish an optimal approach for low-resource settings. Using the nationally representative FinScope Uganda 2023 dataset (survey responses = 3176), we engineered features from UTAUT constructs, security behavior indicators and digital access patterns. A binarized proxy for MFA adoption was derived from validated, high-loading security perception items. Methodologically, we implemented an experimental pipeline involving stratified train–test splitting, SMOTE applied within each cross-validation fold to avoid data leakage and repeated (n = 30) experiments to ensure stability of estimates. Six predictive models, Logistic Regression, Random Forest, Gradient Boosting and XGBoost alongside custom-built Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) architectures, were trained and optimized. Gradient Boosting achieved the strongest performance (mean accuracy = 0.838; F1-score = 0.835; AUC-ROC = 0.928), outperforming both linear baselines and complex neural models, which struggled with recall and F1-scores on tabular survey data. Timing analysis showed that Gradient Boosting balanced computational efficiency with predictive accuracy, making it suitable for low-bandwidth, resource-constrained environments. SHAP-based interpretability revealed that trust in digital security, prior exposure to mobile services, perceived effort and peer influence were the most influential drivers of MFA adoption. The findings advance current knowledge by integrating UTAUT constructs with explainable AI, strengthening behavioral prediction models in sub-Saharan Africa, where empirical MFA studies remain limited. This study contributes a reproducible, theory-grounded modeling pipeline, detailed comparative analysis between classical and neural network approaches and evidence-based policy recommendations.
Prevalence, associated factors, and cutaneous effects of skin-lightening practices among patients at Mbarara Regional Referral Hospital Skin Clinic, Uganda: A cross-sectional study
(JAAD international, 2026) Simon Okongo; Stephen Kizito Mirembe; Grace Mulyowa; Gladys Aloyo
Skin-lightening is a widespread practice among non-White populations, involving the use of soaps, creams, and powders with agents like corticosteroids, hydroquinone, and mercury to reduce melanin and lighten skin tone. However, these products often lead to harmful skin effects. Many people with skin of color lighten their skin for beauty and desirability, influenced by entertainment portrayals favoring lighter tones. Post inflammatory hyperpigmentation from skin disorders also drives self-medication with topical creams. In Uganda, weak regulation makes such products easily accessible, possibly contributing to a rise in skin lightening practices.
Feasibility and Effectiveness of a Simulation-Based Flipped Classroom for Teaching Surgical Suturing Skills in a Low-Resource Setting
(Journal of Medical Education and Curricular Development, 2026) Kirya Musa; Josephine Namugenyi; Charity Mutesi; Kyegombe Willy; Paul Otim; Fred Kirya; Wilfred Arubaku
Background: Surgical skills training in low-resource settings, such as Uganda, often relies on traditional didactic methods with limited hands-on experience, resulting in competence gaps among graduates. The flipped classroom model, a student centered approach where foundational content is learned outside the classroom, and in-class time is used for active practice, offers a promising alternative, particularly for skill-based instruction. This study explored the feasibility and effectiveness of a simulation-based flipped classroom in teaching surgical suturing skills to third-year medical students at Soroti University. Methods: This was a quasi-experimental study in which 55 third-year students participated in a flipped learning module that incorporated pre-class materials (videos, readings, and suturing kits) and in-class simulations using porcine skin. Outcomes were measured through pre-, post-, and 3-month follow-up multiple-choice tests for knowledge, as well as Objective Structured Assessment of Technical Skills (OSATS) scores for skill performance. Results: Post-test knowledge scores showed a marked improvement compared to pre-test scores (mean gain = 39.9, P < 0.0001), with a normalization gain of 70%. In the OSATS observation checklist, most students were rated as competent (49%) or excellent (33%) immediately after training. At 3 months, knowledge retention was 88% despite a statistically significant decline (P < 0.0001). Interestingly, OSATS performance scores improved at 3-month follow-up (+3 points, P=0.0011), suggesting procedural consolidation over time. Conclusion: The flipped simulation model appears feasible and may be effective in enhancing early surgical skill, and knowledge acquisition and retention among medical students in low-resource settings. It fosters experiential learning and improves both cognitive and psychomotor competence. These findings support the potential for expansion to other skills and subsequent integration of this approach into undergraduate surgical training curricula in LMICs.
Modeling the effect of alcohol consumption on the progression dynamics of Hepatitis C infection with treatment
(Mathematics in Medical and Life Sciences, 2026) Agnes Mbeine Asiimwe; Michael Byamukama; Martin Karuhanga
This research introduces a model that integrates a logistic approach to alcohol consumption to delve into the effect of alcohol consumption on the advancement of hepatitis C infection, evaluates the effects of addressing hepatitis C treatment and reducing alcohol consumption to minimal levels in the control of hepatitis C infection. The hepatitis C model was well established and biologically reasonable, indicating that all model solutions were non-negative and associated with non-negative initial conditions. Stability analysis revealed that the disease-free and endemic equilibrium points were locally and globally asymptotically stable whenever the basic reproduction number, R0H ≤ 1 and R0H > 1 respectively. Sensitivity and numerical analysis of the model showed that improving treatment for individuals with chronic infection, along with reducing alcohol consumption, led to a notable decrease in the number of people affected by chronic infection. PLAIN LANGUAGE SUMMARY Hepatitis C (HCV) remains a major global health challenge. This study uses a mathematical model to explore how alcohol use and medical treatment affect the spread and progression of the disease. The model shows two possible outcomes: either the disease fades away or it continues to spread. Which outcome occurs depends on a key number called the basic reproduction number. If this number is less than one, each infected person spreads the virus to fewer than one other person on average, and the disease eventually dies out. But if the number is greater than one, the disease can keep spreading in the population. Computer simulations based on the model suggest that increasing access to treatment for people with chronic HCV and helping them reduce alcohol use can greatly reduce the number of long-term infections. Reducing alcohol consumption not only helps protect the liver but also makes medical treatment more effective.
Assessing the validity of post-discharge readmission and mortality as a composite outcome among newborns in Uganda
(PloS one, 2026) Abhroneel Ghosh; Vuong Nguyen; J. Mark Ansermino; Yashodani Pillay; Angella Namala; Joseph Ngonzi; Nathan Kenya-Mugisha; Niranjan Kissoon; Matthew O. Wiens
Background: Composite outcomes, which include mortality and readmission rates, are often used in risk prediction models following hospital discharge when event rates for the primary outcome of interest, mortality, are low. However, greater readmission rates may result in reduced mortality making interpretation of the composite outcome difficult. We assess the usefulness of a composite outcome of post-discharge readmission and mortality as a target outcome in this context. Methods: This was a secondary analysis of data collected among mothers and their newborn(s) admitted for delivery at two regional referral hospitals in Uganda. Six-week post discharge mortality (all-cause) and readmission in newborn infants were analyzed using a competing risk framework. The Sub Distribution Hazard Ratios (SHRs) were compared across predictor variables to examine the relationship between the two outcomes. Results: A total of 6040 newborns with complete six-week follow-up were enrolled, of whom 50.6% were male and 64% of mothers delivered via caesarean section. Thirty-five (0.58%) infants died within the six-week follow-up period and 241 (3.99%) were readmitted. Of the 206 predictors, 81 had a consistent association with both outcomes. These include a higher weight (SHRs: 0.14, 0.68) and length of the baby (SHRs: 0.85, 0.91). However, 125 variables depicted an association in opposing directions which may be linked to social and financial barriers to care-seeking. These include a travel time to the hospital of greater than 1 hour (SHRs: 1.4, 0.28). Conclusion: While mortality is unequivocally a negative outcome, readmission may be a positive outcome, reflecting health seeking, or a negative outcome, reflecting recurrent illness. This directional dichotomy is reflected to varying degrees within different variables. When using a composite outcome for a prediction model, caution should be exercised to ensure that the model identifies individuals at risk of the intended outcomes of interest, rather than merely the proxies used to represent those outcomes. Identifying predictors with a consistent relationship for both outcomes may yield a more optimized and less biased prediction model for use in clinical care.
Determinants of underweight and overweight/ obesity among people with tuberculosis in Kampala, Uganda: A cross-sectional study
(PLoS One, 2026) Simon Kyazze; Saidi Appeli; Joseph Baruch Baluku; Jonathan Izudi
Background: Malnutrition significantly contributes to mortality among people with tuberculosis (TB). However, evidence on factors associated with the specific forms of malnutrition, specifically underweight and overweight/obesity, beyond clinical determinants, remains limited in many settings. We investigated the prevalence and determinants of underweight and overweight/obesity among people with pulmonary TB across five health facilities in Kampala, Uganda. Methods: This analytic cross-sectional study involved people with pulmonary TB, either clinically diagnosed or bacteriologically confirmed, aged ≥18 years sampled across five health facilities in Kampala, Uganda. Nutritional status was assessed using body mass index (BMI, kg/m²) and categorized as underweight (<18.5), normal weight (18.5–24.9), and overweight/obese (≥25.0). To identify factors independently associated with nutritional status, normal weight was considered as the reference category in a multinomial logistic regression analysis, adjusting for multiple covariates and clustering at the health facility level. The measure of association was the adjusted relative risk ratios (aRRR) and the corresponding 95% confidence intervals (CI). Results: Of the 818 participants studied, 417 (51.0%) had normal weight, 302 (36.9%) were underweight, and 99 (12.1%) were overweight or obese. Adjusted analysis showed that being underweight was associated with household food insecurity (aRRR 2.04, 95% CI: 1.48–2.80) while being overweight or obese was associated with selfemployment (aRRR 2.26, 95% CI: 1.35–3.79) and being newly diagnosed with TB (aRRR 2.10, 95% CI: 1.30–3.41). Conclusion: This study, conducted among people with pulmonary TB in an urban setting in Uganda, showed that underweight and overweight/obesity are prevalent. Furthermore, the study showed that food insecurity is associated with being underweight, while being overweight or obese is associated with being self-employed or newly diagnosed with TB. Therefore, TB control programs need to regularly assess the nutritional status of people with TB to mitigate the effect of being underweight or overweight on treatment outcomes.
Intention to disclose medical errors by healthcare providers to patients and associated factors in Ugandan health facilities: A cross-sectional study
(PLOS Global Public Health, 2026) Catherine Alupo; Jonathan Izudi; Christopher Ddamulira
This study assessed the prevalence and factors associated with intention to disclose medical errors by healthcare providers to patients in public health facilities in Mukono District, Uganda. We conducted an analytic cross-sectional study among healthcare providers across four public health facilities in Mukono District, Uganda. The outcome, Intention to disclose medical errors to patients, was defined as self-reported likelihood of informing a patient or their caregiver about a medical error if it had occurred, including the circumstances of error occurrence and any corrective actions taken. Data were collected using a self-administered structured questionnaire capturing factors at the individual and institutional levels. Logistic regression analysis was used to identify factors independently associated with intention to disclose medical errors, with adjusted odds ratios (aOR) and 95% confidence intervals (CI) reported. We studied 166 healthcare providers, of whom 90 (54.2%) were aged ≥35 years and 95 (57.2%) were female. Overall, 74 (44.6%) participants reported an intention to disclose medical errors. After adjusting for potential confounders, nurses (aOR 0.09, 95% CI 0.02-0.37) and other cadres (aOR 0.09, 95% CI 0.02-0.47) had significantly lower odds of intention to disclose medical errors compared with medical doctors/ physicians. In contrast, healthcare providers working in health facilities with supportive supervision had higher odds of intention to disclose medical errors than those without supportive supervision (aOR 3.32, 95% CI 1.37-8.06). Fewer than half of healthcare providers have the intention to disclose medical errors to patients, indicating a substantial gap from the ethical expectation of full transparency. Supportive supervision was associated with an increased likelihood of intention to disclose medical errors, while non-physician cadres were less likely to report intention to disclose medical errors compared to medical doctors. Interventions that strengthen supervision and empower all cadres through training and institutional support may enhance openness and accountability in healthcare practice.