Background: Tuberculosis (TB) remains a major public health challenge in Somalia, where fragile health systems, delayed care-seeking, limited health literacy, and stigma may undermine TB prevention and treatment. This study assessed TB awareness, treatment-related beliefs, care-seeking intentions, and factors associated with belief in completing the full course of TB treatment among households in Mogadishu. Methods: A community-based cross-sectional study was conducted among 403 adult household respondents in Mogadishu, Somalia. Data were collected using a structured questionnaire covering sociodemographic characteristics, TB awareness, treatment-related perceptions, care-seeking intentions, and household preventive practices. Descriptive statistics were used to summarize participant responses. Bivariate and multivariable logistic regression analyses were performed to identify factors associated with belief in completing the full course of TB treatment. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported, with statistical significance set at p < 0.05. Results: Most respondents had heard of TB (85.4%), recognized coughing or sneezing as a mode of transmission (87.1%), and identified persistent cough as a symptom (86.4%). Overall, 83.1% believed in completing the full course of TB treatment, while only 48.6% correctly identified bacteria or germs as the cause of TB and 49.4% reported discomfort living with a person with TB. In the adjusted analysis, treatment completion belief was positively associated with having no formal education (AOR = 7.91, 95% CI: 2.27–27.54), recognizing persistent cough as a TB symptom (AOR = 15.03, 95% CI: 4.02–56.20), encouraging others to seek care at a health facility (AOR = 12.97, 95% CI: 3.65–46.12), and being comfortable living with a person with TB (AOR = 3.84, 95% CI: 1.38–10.64). Believing that TB is curable when diagnosed early showed an inverse adjusted association (AOR = 0.24, 95% CI: 0.06–0.87). Conclusions: Although TB awareness and treatment completion beliefs were generally high, important gaps in biomedical knowledge and stigma-related perceptions remained. Community-based TB education should strengthen the understanding of disease causation, symptom recognition, timely healthcare seeking, treatment completion, and supportive attitudes toward people affected by TB. The unexpected associations observed in the adjusted model warrant cautious interpretation and further investigation.
Hospital notification of dengue cases is frequently delayed, contributing to underreporting. At Thai provincial hospitals the bottleneck is administrative, not diagnostic: delayed discharge coding and clerical steps mean notifications are submitted late or omitted. Existing machine learning (ML) studies require clinician-annotated features absent in passive electronic medical records (EMRs). We propose a dual-pathway EMR framework for ML-based pre-population of surveillance reports. We analyzed 1031 dengue-suspected encounters (508 notifiable) from Surin Hospital, Thailand (2019–2022). Pathway A combines EMR data with manually documented clinical features; Pathway B uses only passive EMR data. Using an 80/20 stratified split, we tested 7 ML methods across 11 feature sets (77 combinations) with SHAP-guided feature elimination and 10-fold cross-validation. On the primary split, Pathway A (CatBoost) achieved an F1 score of 96.62%, 95.10% sensitivity, and 98.10% specificity (10-seed mean F1 95.7% ± 1.4%). Pathway B (LightGBM) achieved an F1 score of 84.42%, 94.12% sensitivity, 75.24% specificity, and a projected negative predictive value of 99.1% at 10% prevalence; 10-seed performance was lower (F1 78.5% ± 4.3%; sensitivity 89.1%). An exploratory LSTM-16 stacking model (F1 83.99%; precision 79.49%) performed best in precision and specificity among Pathway B. Test F1 SD across 10 seeds was ≤5.0% in 75 of 77 combinations. Within this cohort, passive EMR modeling achieved 91–94% sensitivity on the primary split (85–89% across seeds) without additional data collection, supporting the feasibility of ML-based notification in resource-constrained hospitals. Whether this reduces dengue underreporting must be established in prospective, multi-center studies.
Tuberculosis (TB) patients with comorbidities may have altered rifampicin pharmacokinetics (PK), particularly in sub-Saharan Africa. This can affect treatment response and suggests a role for therapeutic drug monitoring (TDM) in identifying suboptimal exposure. We describe the rifampicin PK in Tanzanian TB patients with HIV, diabetes mellitus (DM), or malnutrition. Participants were enrolled and followed for six months, with PK sampling at baseline and at week 2. All received standard rifampicin, isoniazid, pyrazinamide, and ethambutol. Plasma concentrations were measured via high-performance liquid chromatography (HPLC) before dosing and at 2 ± 1, 5 ± 1, and 12 ± 3 h post-dose. The area under the concentration–time curve to the last measured concentration (AUC0-last) and the peak plasma concentration (Cmax) were derived by non-compartmental analysis. Treatment outcomes were assessed at 6 months using WHO definitions. Of 40 participants (median age 45 years), 34 had a valid PK sample at both visits. Exposure declined between visits in most participants, but magnitude and direction change differed by comorbidity group. The largest median decline occurred in combined HIV–malnutrition (−23.7% for Cmax, and −29.8% AUC0-last), whereas the malnutrition-only group showed no consistent direction, and some of its participants had higher exposure at week 2. In the adjusted mixed-effects model, visit was the only variable whose confidence interval excluded zero (AUC0-last −22.6%, 95% CI, −28.5%, to −16.2%; Cmax −20.1%, 95% CI, −25.5%, −14.3%); HIV, malnutrition, their interaction, age, sex, and mid-upper arm circumference (MUAC) showed no clear association with exposure. Six-month outcomes differed by comorbidity, with deaths concentrated among HIV-positive participants, particularly those who were also malnourished. These findings support combining TDM with individualized dosing, including model-informed precision dosing delivered through clinical decision support systems to optimize rifampicin exposure in high-risk groups.
Background: Molecular assays for the Mycobacterium tuberculosis complex (MTBC) are validated and authorized separately for human and veterinary use, although they target conserved sequences of the same pathogen. This separation, driven by regulatory rather than biological reasons, limits the flexible use of available diagnostics, particularly in low- and middle-income countries. We assessed whether a human-intended and a veterinary-intended MTBC real-time PCR kit retain their ability to detect MTBC in clinical samples from their non-target host. Methods: Human (n = 108; Xpert MTB/RIF Ultra-positive) and bovine (n = 100; MGIT culture-positive) clinical samples were tested with the Viasure (human) and ID Gene (veterinary) kits using the same DNA extract. We compared each kit’s sensitivity to the reference method, agreement between kits, and paired Cq values. Results: In human samples, sensitivity was 88.9% (95% CI 81.4–94.1) for ID Gene and 77.8% (68.8–85.2) for Viasure (p = 0.002). All detection failures occurred in 41 samples with longer storage and additional freeze–thaw cycles, with no evidence of PCR inhibition; the remaining 67 samples had sensitivities of 98.5% and 100%, respectively. In bovine samples, both kits showed equivalent sensitivity (74.0% [64.3–82.3] and 75.0% [65.3–83.1]; p = 1.000). Agreement between kits was 87.0% (κ = 0.54) in human and 91.0% (κ = 0.76) in bovine samples. Among samples detected by both kits, ID Gene amplified earlier than Viasure by a similar margin in human and bovine samples (median 2.5 and 2.7 cycles; p < 0.001). Conclusions: Both kits detected MTBC in samples from their non-target host, with a relative performance independent of host species; differences in sensitivity were explained by sample storage and DNA load rather than by host-related factors. These findings support a One Health approach to validating MTBC molecular diagnostics, while specificity across hosts should be confirmed with MTBC-negative samples.
Background: Fascioliasis caused by Fasciola hepatica and F. gigantica is an important foodborne zoonotic trematodiasis affecting ruminants and humans worldwide. Ecuador has long been recognized as an endemic country, particularly in the Andean region; however, available knowledge remains fragmented across medical, veterinary, and malacological fields. Here, we update and synthesize the available evidence on the epidemiology of fascioliasis with the aim of identifying critical knowledge gaps and guiding future research and national policy. Methods: This study was conducted according to the PRISMA-ScR guidelines. Four electronic databases (Web of Science, PubMed, Google Scholar, and SciELO) were systematically searched through July 2026. Additional publications were identified through manual searches of reference lists, university repositories, and grey literature. Eligible publications reporting original data on infection in humans and other herbivorous mammals, environmental contamination, Fasciola species, lymnaeid intermediate host snails, including their presence, geographical distribution, abundance, or infection status in Ecuador, were analysed. Two reviewers independently screened eligible studies and extracted relevant data, which were subsequently synthesized using descriptive analyses and narrative synthesis. Results: Thirty-six studies published between 1995 and 2026 met the inclusion criteria. Most (63.9%) investigations were conducted in the Andean region, where high infection rates in livestock and competent lymnaeid snails sustain endemicity. Livestock fascioliasis was extensively documented in 15 (42%) studies, with prevalence ranging from 0% to 84.8% in cattle, 61.7–100% in sheep, 21.9% in pigs, and 10.7% in guinea pigs. Liver condemnation due to fascioliasis generated substantial economic losses, although the overall national burden remains unknown. Human evidence was limited to one seroepidemiological survey, demonstrating a 6% seroprevalence in one rural Andean community, and three case reports. In 17 studies, five lymnaeid snail species (Galba cousini, G. schirazensis, G. neotropica, G. cubensis, and Pseudosuccinea columella) were reported, with G. cousini showing evidence as the intermediate host. Molecular studies identified F. hepatica as the only species circulating in Ecuador. No study was found on plant carriers, and none was conducted in the Galapagos Islands. Triclabendazole (TCBZ) was effective in sheep and cattle, with high reinfection rates in 5 months; no study reported on TCBZ resistance. TCBZ is unavailable for human use. There is an absence of a national control programme, standardized diagnostic algorithms or contemporary human epidemiological studies. Conclusions: Fascioliasis remains an important yet neglected zoonosis in Ecuador, with active transmission in the Andean region where infected livestock, competent lymnaeid snails, and suitable environmental conditions coexist. F. hepatica is the sole circulating species and G. cousini the intermediate host; the epidemiological role of other lymnaeid and animal species requires further investigation. Coordinated One Health studies, equitable access to diagnostics and triclabendazole, standardised molecular monitoring, and integrated human–animal-environmental control strategies should be prioritised to reduce Fasciola transmission in Ecuador.
