Julius Kiwanuka

Background: Strengthening hepatitis B(HBV) linkage and retention in care remains critical to the elimination of population health risks posed by HBV. Yet too few studies have assessed the effectiveness of peer-support-driven HBV care models in enhancing continuity in care in low-resource settings. Methods: We co-designed and evaluated the impact of a treatment peer-support (TPS) intervention on HBV linkage and retention in care at two study sites (Arua and Koboko Hospitals) located in the north-west sub-region of Uganda using a quasi-experimental interrupted time series design and assessed factors related to long-term retention. Results: After TPS intervention, changes in the post-intervention trends of linkage to care of 16.2% (95% CI: 3.3, 29.0) (p = 0.014) and 45.9% (95% CI: 26.1, 65.6) (p < 0.001) were observed for Arua and Koboko respectively. Differences between pre- and post-intervention trends were 11.9% (95% CI: 0.3, 23.5) (p = 0.044) and 34.6% (95% CI: 29.5, 39.7) (p < 0.001) for Arua and Koboko respectively. The post-intervention level change for early retention increased by 6.4% (95% CI: 0.4, 12.4) (p = 0.035) at Arua but not at Koboko (22.8%, 95% CI: −20.9, 66.5) (p = 0.306). At Arua, the TPS intervention correlated with an overall declining trend of −3.9% (95% CI: −5.4, −2.3) (p < 0.001), while at Koboko it correlated with a difference between pre- and post-intervention trends of 0.7% (95% CI: −0.01, 1.4) (p = 0.053). For long term retention, the TPS intervention correlated with a change in post-intervention trend of −1.5% (95% CI: −2.3, −0.7) (p < 0.001) and a difference between pre- and post-intervention trends of −1.4% (95% CI: −2.1, −0.6) (p < 0.001) at Arua hospital. At Koboko hospital, the TPS intervention was associated with a difference between pre- and post-intervention trends of 12.8% (95% CI: 9.8, 15.8) (p < 0.001). In the stratified analysis, early retention was associated with the TPS intervention at Koboko (aPR = 1.16, 95% CI: 1.05, 1.28) (p = 0.004) but not Arua (aPR = 1.01, 95% CI: 0.95, 1.06) (p = 0.771). At Arua hospital, a previous history of HBV treatment predicted early retention in care (aPR = 1.08, 95% CI: 1.05, 1.11) (p < 0.001), whereas at Koboko, a PLHBV having a “treatment buddy” (aPR = 1.21, 95% CI: 1.11–1.33) (p < 0.001), being aged 19–24 years (aPR = 1.39, 95% CI: 1.04, 1.88) (p = 0.026) and being aged 25–34 years (aPR = 1.38, 95% CI: 1.04, 1.83) (p = 0.023) predicted early retention. Conclusions: Innovative interventions are needed to advance linkage to care and shift the trajectory of retention in long-term HBV care, including approaches that target young people living with HBV.

Hepatitis B virus (HBV) mother‐to‐child transmission (MTCT) remains a major challenge in Uganda, where 43% of births occur at home, and access to timely birth dose (BD) vaccination is limited. We evaluated the impact of combining universal three‐dose (3D) infant vaccination, targeted BD vaccination, and maternal antiviral treatment for high viral load (HVL) mothers on HBV prevalence among infants. In a prospective 18‐month cohort study at five regional maternity hospitals, HIV‐negative pregnant women were screened for HBsAg. HBsAg‐positive mothers were stratified by HBV DNA into HVL (≥ 20,000 IU/mL), low viral load (LVL), or LVL on treatment. HVL mothers received Tenofovir Disoproxil Fumarate plus Lamivudine in the third trimester. Infants of HBsAg‐positive mothers received BD vaccination within 24 h and the 3D pentavalent series; infants of HBsAg‐negative mothers received the 3D series only. All infants were tested for HBsAg at nine months. Of 19,053 mothers screened, 3.46% were HBsAg‐positive ( n = 660). Among HVL mothers ( n = 80), 96.3% initiated antivirals; none of their infants who received both BD and 3D tested HBsAg‐positive. Among LVL mothers not on treatment ( n = 558), the prevalence of infection at nine months was 0.4% despite full vaccination. No infections occurred among LVL mothers on treatment or among infants of HBsAg‐negative mothers. Overall infant HBsAg prevalence was 0.19%, with an adjusted national estimate of 0.02%. Integrating maternal antiviral prophylaxis, targeted BD vaccination, and universal 3D vaccination achieved near‐elimination of HBV MTCT in this setting. Nationally scaling this strategy could enable Uganda to meet the WHO target of less than 0.1% infant HBV prevalence before 2030 and serve as a model for other high‐burden countries with significant rates of home births.

Introduction: Adolescents and young adults (AYAs) worldwide display poor dietary behaviors, including high consumption of sugar-sweetened beverages and insufficient intake of fruits and vegetables. These issues are more pronounced in Sub-Saharan Africa, such as rural Eastern Uganda, where 45.3% of adolescents eat low-diversity diets high in refined grains and fats. Such diets raise the risk of malnutrition and diet-related non-communicable diseases (NCDs). However, there is limited contextual evidence on the multi-level factors influencing AYAs’ dietary behaviors in rural Uganda. This study examined socio-ecological factors shaping dietary behaviors among AYAs in this setting. Methods: A qualitative study guided by the socio-ecological model (SEM) was conducted in Mayuge District, Eastern Uganda. Focus group discussions (FGDs) were held with AYAs, including male and female, aged 10–14, 15–19, and 20–24 years. To have a nuanced understanding of how AYAs’ dietary behaviours are shaped, additional FGDs were conducted with parents or guardians and teachers, and key informant interviews with the district education officer, nutrition focal person, civil society staff, and food vendors. Data were analyzed in ATLAS.ti using both inductive and deductive thematic approaches: data-driven sub-themes were first identified inductively, then deductively mapped onto pre-determined themes of the SEM. Results: Dietary behaviors were shaped by satiety, energy needs, sensory appeal, and nutrition knowledge at the individual level. Peer influence, parental control, and food’s perceived link to attractiveness acted interpersonally, while community factors included gendered cultural taboos, norms, and health worker advice. At the societal level, cultural identity, ancestral restrictions, and media exposure strongly influenced choices. Conclusions: This study contributes novel rural-specific evidence from rural Uganda, where AYAs’ diets are uniquely constrained by satiety demands, parental dominance, cultural taboos, and seasonal scarcity; contrasting with urban contexts where convenience, autonomy, and wider food environments prevail. Multi-level interventions integrating nutrition education, family and peer engagement, cultural dialogue, and household food security support are essential for promoting healthier diets in resource-limited rural settings.

Background: Adolescent malnutrition, including stunting, underweight, and micronutrient deficiency, is a major public health concern in Sub-Saharan Africa (SSA). While studies from the global north and Asia have shown that nutrition literacy supports healthier diets, evidence on literacy status and its role on influencing overall diet quality among adolescents and young adults (AYA) in SSA is limited. This study assessed nutrition literacy and its association with diet quality among AYA in rural Uganda. Method: This was a cross-sectional study implemented as part of ARISE-NUTRINT project. Based on Nutbeam’s model of health literacy, the study was conducted among 1206 AYA aged 10–24 years in Mayuge district, Eastern Uganda, selected through stratified random sampling. Using structured questionnaire, the Global Diet Quality Score (GDQS) was adapted to estimate diet quality and Adolescent nutrition literacy scale to assess nutrition literacy status. Logistic regression models were employed to assess the association between nutrition literacy and diet quality, adjusting for socio-demographic characteristics. Results: Among 1206 respondents (51.1% female), 85.9% were still in school, over 62% were from low social economic status households, and only 14% used mobile phones. Low nutrition literacy (49%) was prevalent, many unfamiliar with balanced diet or ignoring dietary advice, although 62% were willing to promote healthy eating. Overall, 12.6% had poor-diet quality based on GDQS, marked by frequent refined grain consumption and low fruits/vegetable intake. Having low nutrition literacy was associated with close to five-fold increase in poor diets (adjusted OR = 4.71, 95% CI: 2.19–10.16) while mobile phone use was associated with better diet quality by 56% (aOR = 0.44, 95% CI: 0.27–0.74). Conclusion: Low nutrition literacy is a significant burden among AYA, and is strongly associated with suboptimal diet quality. Implement targeted interventions for improving nutrition literacy can enhance diet quality among AYA in the study area.

Introduction: In Western countries, right-sided colon cancers (RSCC) present at an older age and advanced stage. Researchers believe that there is a difference between left-sided colon cancer (LSCC) and RSCC. In Uganda, however, it is unknown whether differences exist in the pathological profile between RSCC and LSCC. The aim of this study was to determine the differences in clinicopathological characteristics between RSCC and LSCC in Ugandan patients. Methodology: A cross-sectional study was conducted in which colorectal adenocarcinoma formalin-fixed paraffin-embedded tissue (FFPE) blocks were obtained from 2008 to 2021. Colorectal specimens were obtained from prospectively recruited patients. In the retrospective study arm, FFPE blocks and data were obtained from the archives of pathology laboratory repositories. Parameters studied included age, sex, location of the tumour, grade, stage, lymphovascular (LVI) status, and histopathological subtype between LSCC and RSCC. Results: Patients with RSCC were not older than those with LSCC (mean age, 56.3 years vs 53.5 years; p = 0.170). There was no difference in the stage between RSCC and LSCC. Poorly differentiated tumours were more commonly found in RSCC than in LSCC (18.7% vs 10.1%; p = 0.038). Moderately and poorly differentiated colonic tumours were more common with RSCC (89.3%) than with LSCC (75.1%) (p = 0.007). Younger patients had more poorly differentiated tumours than older patients (19.6% versus 8.6%; p = 0.002). LVI was more common with RSCC than with LSCC (96.8% vs 85.3%; p = 0.014). Mucinous adenocarcinoma (MAC) was more common with RSCC (15.8%) compared with LSCC (8.5%) (p = 0.056) although statistical significance was borderline. Conclusions: Clinicopathological features of RSCCs tend to be different from those of LSCCs. RSCCs tend to be associated with MAC, a higher grade and LVI status compared to LSCC. LSCC and RSCC present predominantly with an advanced stage; therefore, national screening programmes for the early detection of CRC are necessary to reduce mortality in our Ugandan population.

Uvulitis is the inflammation and swelling of the uvula, usually associated with infection of nearby structures. Uvulitis can be treated symptomatically, using medication or in some cases with uvulectomy, the uvula surgical removal or shortening. Traditional uvulectomy by traditional practitioners has been practiced in Africa for ages, associated with adverse outcomes. Although there is no empirical evidence for the association between adverse outcomes and traditional uvulectomy in Uganda, anecdotal findings showed incidents of uvula infections following uvulectomy in central Uganda. While these findings also indicate that traditional uvulectomy is common, the community understanding of uvulitis, the beliefs and practices are not well understood. This qualitative study sought to understand beliefs and practices using interviews with community health workers, traditional uvulectomy clients, and traditional surgeons, and focus group discussions with community members. Transcribed data were analysed in Atlas.ti 9 using thematic analysis steps. The findings show that uvula infection, locally known as “Akamiro” and the associated traditional uvulectomy are common in Luwero and beyond. “Akamiro” was described as larger than the normal, the size of a chicken heart or a big pimple, visible when a child cries, with unknown causes. Symptoms included persistent cough, diarrhoea, vomiting, loss of appetite, inability to swallow and ultimately weight loss, swollen stomach, saliva overflow, fever, breathing and speech difficulty. Diagnosis was confirmed after seeking care from health workers or in consultation with significant others and finally the traditional surgeon; in a hierarchical pattern. Uvulectomy was conducted by traditional surgeons, with surgery lasting a few minutes, in the morning or after sun-set. Tools used were razor blades, reeds, strings, wires, sickle knives and spoons. Payment was flexible; cash or in-kind. Surgeons had immense community trust, including community health workers. Interventions to support persons with uvula infections need to address the health system weaknesses, and health education.

Introduction: The incidence of colorectal cancer (CRC) is increasing in East Africa. Changes in lifestyle and dietary changes, particularly alcohol consumption, smoking, and consumption of cooked meats with a reduction in fibre in the diet may be responsible. The objective of our study was to determine the risk factors responsible for CRC in Uganda. Methods: We recruited 129 participants with histologically proven colorectal adenocarcinoma and 258 control participants from four specialized hospitals in central Uganda from 2019 to 2021. Controls were block matched for age (± 5 years) and sex of the case participants. The risk factor variables included; area of residence, tribe, body mass index (BMI), smoking, alcohol consumption and family history of gastrointestinal cancer. We used conditional or ordinal logistic regression to obtain crude and adjusted odds ratios for risk factors associated with CRC. Results: In bivariate analysis, case participants were more likely to be associated with urban residence (cOR:62.11; p< 0.001); family history of GI cancer (cOR: 14.34; p=0.001); past smokers (cOR: 2.10; p=0.080); past alcohol drinkers (cOR: 2.35; p=0.012); current alcohol drinkers (cOR: 3.55; p< 0.001); high BMI 25– 29.9 kg/m² (cOR: 2.49; p< 0.001); and high BMI ≥ 30kg/m² (cOR: 2.37; p=0.012). In the multivariate analysis, urban residence (aOR: 82.79; p< 0.001), family history of GI cancer (aOR: 61.09; p< 0.001) and past smoking (aOR: 4.73; p=0.036) were independently associated with a higher risk of developing CRC. Conclusion: A family history of gastrointestinal cancer was a risk factor for CRC. While population-based CRC screening may not be feasible in low income-countries, targeted CRC screening for first-degree relatives with CRC should be considered in East Africa. Molecular genetic studies need to be carried out to determine the role of hereditary factors in our population. Prevention strategies should be adopted to avoid smoking in our population which was associated with an increased risk of CRC.

Background: Cryptococcal meningitis (CCM) remains one of the leading causes of mortality among HIV infected patients. Due to factors such as the severity of CCM pathology, the quality of life (QOL) of patients post-treatment is likely to be poor. Few studies have reported on QOL of CCM patients post treatment completion. We used data collected among patients in the CryptoDex trial (ISRCTN59144167) to determine QOL and associated factors at week 10 and six months from treatment initiation. Methodology: CryptoDex was a double-blind placebo-controlled trial of adjunctive dexamethasone in HIV infected adults with CCM, conducted between 2013 and 2015 in six countries in Asia and Africa. QOL was determined using the descriptive and Visual Analog Scales (VAS) of the EuroQol Five-Dimension-Three-Level (EQ-5D-3L) tool. We derived index scores, and described these and the VAS scores at 10 weeks and 6 months; and used linear regression to determine the relationship between various characteristics and VAS scores at both time points. VAS scores were interpreted as very good (81–100), good (51–80), normal (31–50) and bad/very bad (0–30). Results: Of 451 patients enrolled in the trial, 238 had QOL evaluations at week 10. At baseline, their mean age (SD) was 35.2(8.5) years. The mean index scores (SD) were 0.785(0.2) and 0.619(0.4) among African and Asian patients respectively at week 10, and 0.879(0.2) and 0.731(0.4) among African and Asian patients respectively at month six. The overall mean VAS score (SD) at 10 weeks was 57.2 (29.7), increasing significantly to 72(27.4) at month six (p<0.001). At week 10, higher VAS score was associated with greater weight (p = 0.007) and being African (p<0.001), while lower VAS score was associated with positive yeast culture at day 14 (p = 0.026). At month six, higher VAS score remained associated with African origin (p = 0.006) while lower VAS score was associated with positive yeast culture (p = 0.006). Lower VAS scores were associated with higher number of inpatient days at 10 weeks and 6 months (p = 0.003 and 0.002 respectively). Conclusion: QOL was good among patients that had completed therapy for CCM, but below perfect. Strategies to improve QOL among CCM survivors are required.

Globally, early initiation of antiretroviral therapy for HIV led to a reduction in the estimated mortality from cryptococcal meningitis (CCM) from 624,700 in 2009 to 181,100 in 2014. However, CCM remains one of the leading causes of mortality among HIV infected patients especially in sub-Saharan Africa where 75% of the deaths occur. Most of the studies evaluating mortality have reported short-term mortality (at or before 10 weeks of therapy). We determined mortality and associated factors among patients treated for CCM in the CryptoDex trial (ISRCTN59144167) in Uganda, and the effect of dexamethasone adjunctive therapy on mortality at two years. We conducted a retrospective cohort study between May 2017 and July 2017 to determine the long term survival (up to 2 years post-randomization) of all patients who had been enrolled into the CryptoDex trial in Uganda. The CryptoDex trial recruited between April 2013 and February 2015. We estimated mortality rates and determined factors affecting mortality at two years using Cox regression. The study followed up 211 participants, 127 (60.2%) of whom were male. Sixteen participants (7.58%) were diagnosed with HIV at the same admission when CCM was diagnosed. By two years following randomization 127 (60%) participants had died, a mortality rate of 67 deaths per 100 person-years. Mortality was associated with Glasgow coma score (GCS) below 15 (adjusted Hazard ratio (aHR) 1.77, 95% CI: 1.02–2.44), p = 0.040; weight (aHR 0.97, per 1 Kg increase; 95% CI: 0.94–0.99), p = 0.003; and presence of convulsions (aHR 2.31, 95% CI: 1.32–4.04), p = 0.004, while dexamethasone use and fungal burden had no effect. Long-term mortality in CCM patients remains high even among patients receiving recommended therapy. Strategies to improve long-term survival in CCM patients are urgently needed, especially targeting those with reduced GCS, low weight, and convulsions.

Retaining patients starting antiretroviral therapy (ART) and ensuring good adherence remain cornerstone of long-term viral suppression. In this era of test and treat (T&T) policy, ensuring that patients starting ART remain connected to HIV clinics is key to achieve the UNAIDS 90-90-90 targets. Currently, limited studies have evaluated the effect of early ART initiation on loss to follow up in a routine health care delivery setting. We studied the cumulative incidence, incidence rate of loss to follow up (LTFU), and factors associated with LTFU in a primary healthcare clinic that has practiced T&T since 2012. We retrospectively analyzed extracted routine program data on patients who started ART from January 2012 to 4^th July 2016. We defined LTFU as failure of a patient to return to the HIV clinic for at least 90 days from the date of their last appointment. We calculated cumulative incidence, incidence rate and fitted a multivariable Cox proportion hazards regression model to determine factors associated with LTFU. Of the 7,553 patients included in our sample, 3,231 (42.8%) started ART within seven days following HIV diagnosis. There were 1,180 cases of LTFU observed over 15,807.7 person years at risk. The overall incidence rate (IR) of LTFU was 7.5 (95% CI, 7.1–7.9) per 100 person years of observation (pyo). Cumulative incidence of LTFU increased with duration of follow up from 8.9% (95% CI, 8.2–9.6%) at 6 months to 20.2% (95% CI, 19.0–21.4%) at 48 months. Predictors of elevated risk of LTFU were: starting ART within 7 days following HIV diagnosis ((aHR) = 1.69, 95% CI, 1.50–1.91), lack of a telephone set (aHR = 1.52, 95% CI, 1.35–1.71), CD4 cell count of 200–350μ/ml (aHR = 1.21, 95% CI, 1.01–1.45) and baseline WHO clinical stage 3 or 4 (aHR = 1.35, 95% CI, 1.10–1.65). Factors associated with a reduced risk of LTFU were: baseline age ≥25 years (aHR ranging from 0.62, 95% CI, 0.47–0.81 for age group 25–29 years to 0.24, 95% CI, 0.13–0.44 for age group ≥50 years), at least primary education level (aHR ranging from aHR = 0.77, 95% CI, 0.62–0.94 for primary education level to 0.50, 95% CI, 0.34–0.75 for post-secondary education level), and having a BMI ≥ 30 (aHR = 0.28, 95% CI, 0.15–0.51). The risk of loss to follow up increased with time and was higher among patients who started ART within seven days following HIV diagnosis, higher among patients without a telephone set, lower among patients aged ≥ 25 years, lower among patients with at least primary education and lower among patients with BMI of ≥ 30. In this era of T&T, it will be important for HIV programs to initiate and continue enhanced therapeutic education programs that target high risk groups, as well as leveraging on mHealth to improve patients’ retention on ART throughout the cascade of care.