JAC-Antimicrobial Resistance

Background : Antimicrobial resistance (AMR) is a major public health threat globally, with a disproportionate burden in sub-Saharan Africa. Faith-based health facilities provide essential healthcare services to underserved populations, yet data on antimicrobial use in these settings remain limited. Aim : To assess antimicrobial use in Christian Health Association of Malawi health facilities using the World Health Organisation core medicine use indicators. Materials and Methods : A multicentre cross-sectional study was conducted in 29 CHAM health facilities across Malawi between January 2024 and June 2025. Data were collected from facility personnel, inpatient prescriptions, and patient interviews and analysed using descriptive and inferential statistics. Results : Average availability of key antimicrobials was 33.1% (95% CI: 29.7-36.4), while customised formularies were available in 64.3% of health facilities. Among 660 prescriptions analysed, 90.3% contained an antimicrobial agent, but only 33.2% adhered to standard treatment guidelines and 43.6% were prescribed using full generic names. Facilities with pharmacy professionals were more likely to have a facility-specific formulary (84.6% vs. 46.7%, p = 0.037). Conclusion : Antimicrobial stewardship gaps remain substantial in faith-based health facilities in Malawi and across sub-Saharan Africa, highlighting the need for targeted stewardship programmes in faith-based health facilities.

Background and objectives Spontaneous bacterial peritonitis (SBP) is a life-threatening complication of cirrhosis with ascites, carrying one- and two-year mortality rates exceeding 70% and 80%, respectively. Fluoroquinolone prophylaxis is the cornerstone of SBP prevention. Real-world longitudinal data on prescribing practices and clinical outcomes from Indian tertiary care centers are sparse. We aimed to evaluate fluoroquinolone prescribing patterns, guideline adherence, and six-month clinical outcomes in SBP patients at a tertiary academic center in North India. Methods This was a pre-specified sub-analysis of a 15-month analytical longitudinal study at AIIMS Rishikesh. Adults (age >/=18 years) admitted with SBP and initiated on fluoroquinolone prophylaxis were enrolled consecutively and followed for six months. Prescribing practices were compared against EASL and AASLD recommendations. The primary outcome was the rate of guideline-directed prescribing. Secondary outcomes included clinical cure at discharge, six-month cure, relapse, regimen modification, adverse drug reactions, and treatment compliance. Categorical variables were compared by Fisher's exact test or chi-squared test (SPSS). Results Forty-eight SBP patients were included (mean age 44.75 +/- 11.94 years; 85.4% male). Guideline-directed fluoroquinolone prophylaxis was prescribed to all patients (100%). Norfloxacin 400 mg once daily was predominant (85.4%), followed by levofloxacin (10.4%) and moxifloxacin (4.2%). Cure at discharge was 85.4%. At six months, 64.6% maintained sustained cure and 22.9% relapsed. Regimen modification occurred in 22.9%, most commonly antimicrobial substitution. Nausea was the only adverse drug reaction (4.8%). Treatment compliance was 73.8%. No patient underwent therapeutic drug monitoring. Conclusions Fluoroquinolone prescribing for SBP prophylaxis at AIIMS Rishikesh was fully concordant with standard guidelines. Despite complete adherence, a relapse rate of 22.9% and frequent regimen modification underscore the limitations of long-term fluoroquinolone prophylaxis, likely reflecting emerging quinolone resistance. Strengthening antimicrobial stewardship is essential to sustain prophylaxis effectiveness in Indian tertiary care settings.

Bacterial contamination of Surgical sites could lead to Surgical Site infections (SSI) which may prolong hospital stay, increased treatment costs and increased mortality. This study determined the prevalence of bacterial contamination of surgical sites among post-operative wounds and theatre surfaces together with their resistance to commonly prescribed antibiotics. A cross-sectional study design was used where a total of 290 and 74 swabs were collected from patients and theatre surfaces respectively. Swabs were cultured on duplicate plates of Blood Agar, Chocolate Agar and MacConkey Agar. Gram-staining and Biochemical tests were performed to identify the isolates. Resistance to commonly prescribed antibiotics was determined using the Kirby Bauer (KB) method. Data were analyzed using SPSS version 23, and descriptive statistics, Chi square and student T- tests were used to describe the results. The prevalence of bacterial contamination in wounds was 30.7% and was significantly higher in women of child bearing age ({chi}2= 10.79, df=1, P=0.0010). Microbial growth increased with an increase in duration of antibiotic therapy ({chi}2=12.73, df=2, P=0.007). E. coli was responsible for the highest cases of wound contamination (34.9%). All microorganisms isolated from post-operative wounds showed considerable resistance to antimicrobials. All isolates from wounds were resitant to Trimethoprin Sulfamexathone and 76.9% showed resistance to Ciprofloxacin. Other than E.coli and Acinetobacter, the rest of the isolates were susceptible to imipinem. Fourty nine gram positive isolates were grown from theatre surfaces and a significant majority (86%) were from air. There was high resistance to Erythromycin in Coagulase Negative Staphylococcus (CNS) isolates (56.0%). Overall, our study demonstrated that wound contamination at the Hoima Regional Referral Hospital is high but not associated with theatre surface contamination.

Background: Antimicrobial resistance (AMR) disproportionately affects low- and middle-income countries (LMICs) such as Pakistan, where obstetric and gynaecological (OBGYN) patients carry high antibiotic exposure. Specialty-specific drug utilization data with concurrent stewardship audit remain scarce. This study evaluated antibiotic prescribing patterns, consumption metrics, and antimicrobial stewardship program (AMS) compliance in OBGYN inpatients at a public sector tertiary care hospital. Methods: A prospective cross-sectional study was conducted in OBGYN wards of Dow University Hospital, Karachi, from 1 September to 31 October 2025. Women receiving [≥]1 systemic antibiotic were included. Daily AMS rounds were conducted by an Infectious Diseases physician and pharmacist. Antibiotic consumption was measured as Defined Daily Doses (DDD) and Days of Therapy (DOT) per 1,000 patient-days (total = 821). Antibiotics were classified by WHO AWaRe (2023) framework. Results: Of 812 total admissions, 278 patients (34.2%) received [≥]1 antibiotic and were enrolled (205 obstetric, 73 gynaecological), generating 636 prescriptions (mean 2.29/patient). Surgical prophylaxis was the predominant documented indication (213, 33.5%); 65.1% carried no documented indication. By AWaRe classification, 53.6% were Access-group and 46.1% Watch-group. Ceftriaxone (38.4%) and metronidazole (36.8%) together represented 75.2% of prescriptions. Combined DDD/1,000 patient-days was 1,758.6 and DOT/1,000 patient-days was 1,852.7. AMS compliance was 0%. Conclusions: This study documents high antibiotic prescribing burden, near-universal documentation failure, and zero AMS compliance in OBGYN inpatients at a Pakistani public sector hospital. The predominance of Watch-group antibiotics and undocumented surgical prophylaxis highlights structural stewardship gaps. Findings support urgent need for institutional OBGYN antibiotic guidelines and structured pharmacist-led AMS programs.

Antimicrobial resistance (AMR) exacerbates bacteraemia in cancer patients, particularly in low-resource settings. At the Uganda Cancer Institute, high rates of Enterobacterales producing extended-spectrum {beta}-lactamases (ESBLs) have been reported, with DNA-based detection of bla genes limited to PCR. This study aimed to determine whether bacterial genomic DNA shipped at ambient temperature from Uganda to the UK retained sufficient quality for whole-genome sequencing (WGS), to allow in-depth genomic analyses of isolates. Genomic DNA was extracted from Gram-negative bloodstream isolates (n=77) in Uganda and shipped to the UK at ambient temperature. rpoB gene (77/77, 100%) and WGS data (72/77, 93.5%) were generated for isolates, with 66/72 (91.7%) genomes of high-quality (Escherichia coli n=34; Klebsiella spp. n=32). Bioinformatic analyses included species identification, sequence typing, SNP analysis, AMR and virulence gene profiling, and comparison with publicly available genomes of Ugandan isolates. Phenotypic-genotypic concordance was generally high: 7/77 (9.1%) isolates were misidentified by phenotypic testing, and two showed unexplained carbapenem resistance. E. coli isolates showed diverse sequence types, with high prevalence of blaCTX-M (91.2%) and blaOXA-1 (47.1%); carbapenemase genes were rare. Klebsiella isolates lacked hypermucoidy loci and displayed diverse capsule types, with a high prevalence of ESBLs. Genomic clustering suggested limited within-hospital transmission of strains. Genomic data can provide important insights into the dissemination of bacterial subclades of global concern. The widespread AMR genotypes reported here highlight the need for improved diagnostics and updated treatment guidelines for bacteraemia in Ugandan cancer patients.

Background Antimicrobial stewardship in many settings assumes that community antimicrobial misuse reflects low awareness, and favours education-based interventions. Population-level evidence on healthcare seeking and antimicrobial practices in marginalised indigenous groups in low-income and middle-income countries is limited. We examined socioeconomic and identity-related determinants of healthcare-provider choice, antimicrobial awareness, and harmful antimicrobial practices in a tribal population in central India. Methods We did a cross-sectional survey of 1146 adults in the catchment area of Jan Swasthya Sahyog (JSS), a non-profit community health organisation, in Bilaspur and Mungeli districts, Chhattisgarh, India (January, 2021-April, 2022). Healthcare-provider choice was modelled with binary and multinomial logistic regression (government as reference), and antimicrobial awareness and ten harmful practices with logistic regression, applying Benjamini-Hochberg false discovery rate (FDR) correction within each family. A 30-day treatment-recall sub-study (n=284) assessed actual treatment location and out-of-pocket cost. Models with rare events or separation were refitted with Firth penalised regression. Findings Median per capita income was INR 8000 per year. Baiga identity was associated with higher odds of using informal (odds ratio 2.58) and private (2.55) providers rather than government facilities, but not JSS (1.36). Awareness of antimicrobials was 7.6% and was associated mainly with education (primary-or-less vs college 0.04). Baiga identity was independently associated with premature discontinuation (4.77), stopping for perceived intolerance (4.43), and financial discontinuation (6.81, 95% CI 3.51-13.25), but with lower odds of stopping because of perceived recovery (0.12). In the sub-study, predicted government-facility use was 1.6% for Baiga versus 18.1% for non-Baiga individuals. Findings were robust to Firth penalisation. Interpretation In this population, antibiotic non-completion was associated with poverty and access constraints rather than only with awareness, and a non-profit provider appeared to reach groups more equitably. Affordability-oriented, differentiated stewardship merits prospective evaluation. Findings are from a single catchment, are associational, and should be interpreted with the study's sampling in mind.

Background: Antimicrobial resistance poses a major threat to global public health. Large language models (LLMs) offer new possibilities for optimizing antibiotic prescribing decisions, but the capabilities of general-purpose versus domain-specific medical LLMs under different prompting strategies remain to be clarified. Methods: This double-blind, randomized-sequence evaluation used a 2X2 factorial design comparing four AI conditions-the domain-specific model MedGo and the general-purpose model DeepSeek V3.5, each under standard direct prompting and chain-of-thought (CoT) prompting-alongside real physician prescriptions across 59 complex inpatient infection cases. Five parallel regimens were generated per case and independently evaluated by three senior clinicians (1-5 comprehensive score and five domain sub-scores). ChatGPT 5.2 was additionally assessed as an automated evaluation tool. Results: Score ranking: real physicians > MedGo-CoT > DeepSeek-CoT > MedGo> DeepSeek (Friedman test, p<0.001). In base mode, MedGo significantly outperformed DeepSeek (Holm-adjusted p=0.040). CoT improved both models (Holm-adjusted p<0.001 for DeepSeek; p=0.024 for MedGo) and reduced score dispersion. MedGo-CoT significantly outperformed DeepSeek-CoT in individualized adjustment (adjusted p<0.001) and dosing precision (adjusted p=0.005). ChatGPT-expert correlation was negligible (overall Kendall {tau}=0.153, p=0.003; subgroup {tau}=0.06-0.20, all p>0.05). Conclusions: Domain-specific medical LLMs enhanced by CoT approach the antibiotic decision-making level of real physicians, with advantages in individualization and dosing precision. However, notable deficiencies persist in antimicrobial stewardship ecological awareness and automated evaluation reliability, underscoring the continued indispensability of senior clinical expertise.

Carbapenem-resistant Klebsiella pneumoniae (CRKp) is a critical global healthcare threat driven by high-risk multidrug-resistant (MDR) clones that acquire hypervirulence genes. Although resistance-virulence co-occurrence is extensively documented, the plasmid-level mechanisms facilitating this convergence remain unclear. In this study, we utilized hybrid short- and long-read whole-genome sequencing of 376 clinical CRKp strains to define the evolutionary trajectories and structural plasmid dynamics of three predominant high-risk clones: ST147 (n=157), ST231 (n=108), and ST2096 (n=111). Carbapenemase genes were present in 90% of isolates, predominantly blaOXA-48-like and blaNDM-5 co-harbored with blaCTX-M-15. Virulence profiling indicated high aerobactin (iuc) prevalence (62.7%), while salmochelin and colibactin were undetected. Hypermucoviscosity occurred infrequently (6.6%) and was independent of rmpA/rmpA2, confirming a clear genotype-phenotype discordance. Comparative plasmid mapping revealed three distinct, lineage-specific plasmid configurations underlying this intermediate convergent pathotype: ST147 exhibited dynamic, mosaic hybrid IncFIB-IncHI1B plasmids; ST2096 showed structurally stabilized hybrids; and ST231 retained virulence and resistance determinants on separate, segregated plasmids. These findings show that convergence is regulated by multiple, clone-specific evolutionary routes rather than a single path, highlighting the critical need for more in-depth genomic surveillance capable of identifying convergent plasmids along with high-risk lineages

Background Klebsiella pneumoniae (Kpn) is a major cause of neonatal sepsis in Africa. 3rd generation cephalosporin and gentamicin resistant Kpn is the norm in many sites, rendering WHO recommended first- and second-line antimicrobials ineffective. An understanding of which neonates and infants are most at risk of sepsis caused by Kpn would support the case for improved access to WHO watch and reserve antimicrobials (i.e. carbapenems) for patients most likely to benefit from them. Methods A prospective case-control study was conducted at Queen Elizabeth Central Hospital, Malawi. Cases were infants <3 months of age with blood or CSF culture confirmed Kpn infection. Controls were healthy infants from the same wards and were matched 2:1. Univariate and multivariate logistic regression were performed on mean-centred data to determine risk factors for infection with Kpn. Results We analysed data from 38 cases and 76 controls between August 2021 and April 2023. Mortality at 3 months of age was 21/38 (29%) for cases, with 14/38 (37%) identified postmortem and 6/76 (7.9%) for controls (OR 14.0 (95% CI 4.59, 49.2, p>0.001). Cases were more likely to be born out of QECH than controls (42% vs. 24%, p = 0.043), and cases had lower birthweights (median 2200g vs. 2850g, p = 0.005). Multivariate logistic regression analysis revealed that increasing birthweight was protective against Kpn infection (OR: 0.858 [95% CI: 0.745, 0.987] per 100g increase), while longer hospital stay was associated with increased odds of infection (OR: 1.148 [95% CI: 1.012, 1.1.303] per additional day). Most infecting isolates (34/38 [89%]) were resistant to first- and second-line antimicrobial agents, but all were sensitive to meropenem and 33/36 [92%] to amikacin. Conclusion Low birthweight infants with prolonged hospital stay were at greatest risk of Kpn infections that were typically resistant to WHO first- and second-line antimicrobial therapy. These infants should be prioritised for antibiotics that have the potential to be life-saving. The overlapping and evolving nature of these risk factors makes it difficult to design a simple tool to support empiric initiation of meropenem. Neonates critically ill with Kpn sepsis cannot, however, afford to wait for blood culture confirmation before receiving effective treatment. This highlights the need for empiric decision making frameworks that allow rapid initiation of effective therapy in high-risk neonates.

Objectives To evaluate two-year, side-by-side outcomes of a prospective audit and feedback (PAF)-based antimicrobial stewardship program (ASP) in a quaternary ICU in Ghana, comparing diagnostic stewardship, antimicrobial prescribing patterns, and clinician adherence to stewardship recommendations between 2024 and 2025. Longitudinal PAF data from low- and middle-income countries (LMIC) quaternary ICUs are scarce; this study addresses that evidence gap. Methods A retrospective comparative analysis of routine Antimicrobial Stewardship (AMS) surveillance data was conducted at the University of Ghana Medical Centre ICU: 102 visits in 2024 and 63 in 2025. Proportions were compared by chi-square or Fishers exact test; continuous variables by Mann-Whitney U. Wilson score 95% confidence intervals (CIs) were computed for primary proportions. Results Biomarker-guided prescribing rose from 86.3% to 100% of visits (p=0.005) and culture and sensitivity testing from 74.5% to 90.5% (p=0.02). Targeted (culture-guided) therapy increased significantly from 23.5% to 41.7% of antibiotic recipients (p=0.03), while empiric prescribing declined correspondingly. Overall antibiotic utilization remained high in both years (96.1% vs 95.2%; p=1.00), and meropenem use rose from 42.9% to 56.7% (p=0.13). AMS interventions were recommended in 67.6% and 63.5% of visits, respectively. Clinician acceptance improved markedly from 40.6% (95% CI: 29.8-52.4%) to 67.5% (95% CI: 52.0-79.9%) (p=0.01). Conclusions Two years of PAF in a Ghanaian quaternary ICU demonstrated progressive program maturation: universal biomarker adoption, a significant shift toward targeted prescribing, and markedly enhanced clinician acceptance. Persistently high antibiotic utilization and rising carbapenem dependence underscore the need for sustained surveillance and carbapenem-sparing strategies in LMIC critical care.

Background Severe Clostridioides difficile infection (CDI) morbidity and mortality disproportionately affect Black and Hispanic patients in the United States. Antibiotic exposure is the primary modifiable risk factor for CDI, and clindamycin is among the agents most strongly associated with related harm. Characterizing inequities in prescribing is critical. Dentistry is a major source of clindamycin prescriptions. Academic dental clinics serve diverse patient populations and provide an ideal setting to evaluate prescribing across racial and ethnic groups. We therefore examined antibiotic use and cumulative clindamycin exposure as measures of CDI-associated risk. Methods We conducted a retrospective study of electronic health records from 5 US academic dental institutions from 2021 through 2023. We analyzed 552,428 encounters among 132,770 patients with documented race/ethnicity to estimate adjusted odds of receiving any oral antibiotic and clindamycin by race/ethnicity. Secondary outcomes evaluated total antibiotic exposure among dental provider-prescribed antibiotics, focusing on higher-than-standard cumulative dosing of clindamycin (>8400 mg) and amoxicillin (>10,500 mg). Results Oral antibiotic prescribing occurred in 1.9% of encounters. Compared with White patients, Black, Hispanic, and Other race patients had slightly lower adjusted odds of receiving any oral antibiotic, while Black patients had greater odds of receiving a higher-than-standard cumulative clindamycin dose when clindamycin was prescribed (adjusted odds ratio, 2.19; 95% confidence interval, 1.25-3.82). Conclusion Racial and ethnic inequities in dental antibiotic prescribing extended beyond antibiotic receipt to cumulative clindamycin exposure. Although CDI outcomes were not directly measured, these prescribing differences may have implications for disparities in CDI-associated harm and warrant further investigation.

Rationale: Infections due to Extended-spectrum {beta}-lactamases-producing Enterobacterales (ESBL-PE) require empirical treatment with carbapenems. ESBL-PE carriage is considered as a risk factor for ESBL-PE involvement during ICU infection. Our aim was to determine factors that may predict the actual involvement of ESBL-PE. Methods: A two-periods bicentric ambispective study including ICU ESBL-PE carriers patients from April 2011 to January 2019. All ESBL-PE carriers who developed an infection were analyzed. Results: 6112 patients and 4902 patients were screened during the two periods. 384 and 232 ESBL-PE carriers were identified. Total number of infectious episodes were 146 and 114, respectively. A total of 144 pneumonias, 42 urinary tract infection and 45 digestive infections were studied. An ESBL-PE was involved in 35 (24.3%) episodes of pneumonia, and 44 (37.9%) of extra-pulmonary infections. The most frequent ESBL-PE involved were K. pneumoniae, E. cloacae and E. coli. Similar species and phenotypes were present in colonisation and infection in 29 (82.8%) of pneumonia and in 40 (90.9%) of extra-respiratory infection. Multivariate analysis identified Klebsiella pneumonia or Enterobacter cloacae carriage as risk factor for ESBL-PE involvement in pneumonia and E. coli carriage and detection of ESBL-PE carriage before ICU admission as protective factors. Conclusion: In our study an ESBL-PE involvement is infrequent in pneumonia. A known carriage before ICU admission and E. coli carriage are factors associated with the absence of ESBL-PE un the episode of respiratory infection. A confirmation of our findings could lead to a reduction in the empirical use of carbapenems in this population.

Background: Urethritis in women of childbearing age constitutes a significant but underreported burden of reproductive morbidity in Sub-Saharan Africa, where diagnostic constraints often necessitate suboptimal syndromic management. Methods: To identify the localized etiological profile, mid-stream urine and urethral swab specimens were prospectively collected from symptomatic women attending local clinics, subjected to standard microbiological culture, and characterized using rigorous phenotypic and biochemical diagnostic protocols. Results: Microbiological analysis successfully isolated a high prevalence of both Gram-negative and Gram-positive uropathogens, predominantly Escherichia coli, Staphylococcus aureus, and Klebsiella pneumoniae, demonstrating distinct phenotypic traits characteristic of the regional microbial ecology. Conclusion: The pronounced isolation of these specific bacterial agents highlights the critical inadequacy of generalized empirical treatments and underscores the urgent need for tailored diagnostic criteria in resource-limited African healthcare settings.

Antimicrobial resistance constitutes a major threat to global public health. Among emerging extensively drug-resistant bacteria (eXDR), carbapenemase-producing Enterobacteriaceae (CPE) expose hospitals to outbreaks through rapid dissemination, and to therapeutic limitations. Through a mixed epidemiological-qualitative methods study, we report the most extensive CPE carriage outbreak known to date in France, which occurred at Hopital Europeen Marseille (HEM) between January and June 2025. By the end of November 2024, the admission of an index patient returning from Senegal carrying an NDM-producing Escherichia coli led to an extensive transmission, despite adherence to national screen and isolate guidelines. More than 7,500 rectal screening tests evidenced 481 CPE carriers (including 343 NDM, 129 OXA-48-like and 9 other CPE), and 14 vancomycin-resistant Enterococcus faecium carriers. This major outbreak conducted to a phenomenal involvement of clinical, technical and administrative teams within the institution. It highlighted operational limitations in current screening, cohorting and biocleaning strategies in the context of hospital-wide outbreak. We describe the outbreak trajectory, the control measures implemented and provide a structured synthesis of lessons learned across organisational, scientific and policy domains.

Background: Carbapenemase producing Enterobacterales (CPE) remain a major infection prevention and control challenge. Although care home residence is frequently perceived as a risk factor for CPE carriage, its independent association with CPE positivity remains uncertain. Objective: To investigate the relationship between care home residence on admission and CPE positivity among patients undergoing CPE screening. Methods: A retrospective matched case control study was conducted at a single NHS acute hospital in England. Adult patients with laboratory confirmed CPE positivity from screening samples between 1 November 2022 and 1 November 2025 were matched to CPE negative controls at a ratio of up to 1:4 based on ward, specimen year and age no more than 5 years older or younger. Conditional logistic regression was used to assess the association between care home residence and CPE positivity. An adjusted model included previous hospital admission within 12 months. Results: A total of 108 CPE positive cases were successfully matched to 412 controls. Care home residence was identified in 14 (13.0%) cases and 49 (11.9%) controls. In the matched conditional logistic regression model, care home residence was not associated with CPE positivity (OR 1.15, 95% CI 0.58 to 2.28; p=0.690) and remained non-significant after adjustment (aOR 1.32, 95% CI 0.66 to 2.64; p=0.439). Discussion: Care home residence was not independently associated with CPE positivity in this low-prevalence setting. Significance and impact: The findings do not support the use of care home residence alone to guide CPE screening. Further multicentre studies are required to clarify the contribution of care home residence to CPE epidemiology.

Motivation: Antimicrobial resistance (AMR) has been identified as a top global public health threat. Accurate AMR phenotype prediction from whole-genome sequencing data is an essential tool for accelerating clinical decision-making and mitigating resistance spread. Although many previous works have explored the use of tree-based machine learning (ML) models to predict resistance, the field lacks a systematic evaluation of the training pipeline across a variety of pathogenic species and antibiotics. Results: Using nine clinically relevant species-antibiotic combinations from the NCBI antimicrobial susceptibility testing database, we present a detailed analysis of the ML pipeline and identify key factors affecting model performance and evaluation. We begin by relabelling all isolates using current CLSI minimum inhibitory concentration breakpoints to resolve inconsistencies and increase available data, resulting in up to a 19% label swap and 56% data enlargement per species-antibiotic combination. We identify several key training parameters including k-mer length, which can increase classification F1 scores by over 20 points compared to commonly used k-values, feature matrix truncation, which can induce polynomial time reductions with limited performance reduction, and ML model class. By comparing 5-fold cross-validation with evaluation on an unseen clinical dataset, we show that random cross-validation splits--often criticized as overly optimistic--can act as a strong proxy for downstream clinical performance, yielding closer F1 scores than phylogeny-aware splits in all cases. We finally present an interpretability study which shows that over 95% of k-mers used by our models are associated with identifiable genomic features. Our results highlight the importance of feature design, evaluation protocol, and biological analysis in genomic AMR prediction, and support tree-based models as a robust and interpretable method.

Antimicrobial resistance genes (ARGs) can spread via horizontal transfer or clonal expansion. We investigated the genomic epidemiology of extended-spectrum beta-lactamase (ESBL)--producing Klebsiella pneumoniae and Escherichia coli in a neonatal unit. Between January and November 2023, 53 ESBL isolates were obtained from 23 neonates via routine screening and clinical sampling. Long-read nanopore sequencing identified blaCTX-M-15 as the dominant ESBL gene, alongside blaCTX-M-65 and blaCTX-M-27. Among 49 blaCTX-M-15 isolates, 33 carried the gene on plasmids and in the remainder it was located on the chromosome. ESBL isolates belonged to one of seven MLST sequence types; E. coli isolates were dominated by ST131/ST131-like/ST5640 lineages, while K. pneumoniae were predominantly ST13. Plasmids (ESBL and non-ESBL-associated) clustered into 18 communities, five of which contained plasmid-bearing blaCTX-M genes. The largest cluster comprised IncFIB(K) plasmids from K. pneumoniae ST13, although these were predicted as "non-mobilizable" by MOB-suite and belonged to isolates from a clonally disseminated strain. No evidence of blaCTX-M dissemination via shared plasmids was identified. Meanwhile, chromosomal phylogenetic analysis identified four distinct clonal clusters with [&le;]7 SNP differences (n = 10, 5, 5, and 2 patients). In this setting, genomic analysis supported clonal dissemination of several blaCTX-M-associated strains as the main outbreak mechanism, affecting 20/23 neonates, rather than plasmid-mediated transmission.

Antimicrobial resistance (AMR) is a global threat to public health and development. Failure to address it could return society to a pre-antibiotic era with increased morbidity and mortality. Because human behaviour is crucial to AMR management, interventions modifying knowledge, attitudes, and practices are therefore essential. Modifying health-related behaviours presents a significant challenge, yet it is crucial for public health. Engaging populations during periods of shifting perceptions can address this challenge and ensure the sustainability of interventions. Adolescents and young people attending school represent a key demographic. The primary objective of this study is to evaluate student's awareness, perceptions, and behaviours concerning antimicrobial resistance (AMR) and hygiene, while enhancing and empowering children as agents of change within the community. In this study, students from Allied Health Sciences (AHS) across various disciplines were recruited to serve as peer educators for an evidence-informed educational workshop. A pilot delivery of these activities was conducted among a few students in a school before the final delivery was executed in three schools. Schools following a comparable educational board and curriculum were selected for inclusion in the study. A structured questionnaire was employed to assess the effects before and after the intervention. Statistically significant improvements were observed in participant's knowledge, attitudes, and practices (p < 0.001). Additionally, feedback was collected from participants, teachers, and the school nurse attending the session. By triangulating these findings, a notable immediate improvement was observed in students' knowledge, attitudes, and practices. This study provides evidence that employing multimodal teaching led by peer education is a valid and effective method for delivering health messages. It further underscores the mutual benefits for stakeholders (peer educators and peer learners) by offering a two-way learning opportunity. The benefits extend beyond academic and core scientific learning to include increased confidence as effective health educators and future-ready healthcare professionals.

Background. The 2024 international reference standard for urinary tract infection (UTI) research scores four domains - symptoms and signs, systemic criteria, pyuria, and culture - to classify samples as No UTI, Possible UTI, Probable UTI, or Definite UTI. It explicitly identifies spinal cord injury (SCI) as a condition of impaired symptom perception, and states that catheter-associated UTI requires a separate standard. We applied it to verified-asymptomatic samples from adults with neurogenic lower urinary tract dysfunction (NLUTD) due to spinal cord injury or disease (SCI/D) who use intermittent catheterization (IC), to test whether it can identify UTI likelihood in this population. Methods. The reference standard was applied to 224 samples from 99 adults with NLUTD due to SCI/D using IC, all verified asymptomatic by the Urinary Symptom Questionnaire for Neurogenic Bladder-Intermittent Catheter (USQNB-IC) at sampling and for 72 hours prior. Because no participant was febrile and no blood markers are drawn in this population, the systemic-criteria domain scored zero for every sample; the reported classifications are therefore a floor. Pyuria was scored under conservative and inclusive interpretations of categorical urinary white blood cell (WBC) bins. Results. Under conservative interpretation, 40.2% of samples were classified No UTI, 21.9% Possible UTI, and 37.9% Probable UTI; under inclusive interpretation, 11.6% No UTI, 38.4% Possible UTI, and 50.0% Probable UTI. No sample reached Definite UTI - a structural consequence of the empty systemic domain. Conclusions. The consensus reference standard classifies 38-50% of fully asymptomatic NLUTD-IC samples as Probable UTI, a floor estimate that could only rise if blood markers were available. This confirms the 2024 framework's own prediction that a separate standard is needed for populations with altered symptom expression and baseline-positive urinary markers.

Background: Tuberculosis, especially drug-resistant tuberculosis (DR-TB) including multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains, remains a leading cause of infectious death worldwide. The rapid accumulation of whole-genome sequencing (WGS) data had spurred numerous computational methods for predicting antimicrobial resistance in Mycobacterium tuberculosis. However, heterogeneous datasets, preprocessing pipelines, and evaluation protocols have made fair comparisons impossible and have hindered clinical translation. A critical yet missing resource is a large-scale, unified benchmark to systematically assess and compare existing methods. Methods: We curated an integrated MTB WGS--phenotypic drug susceptibility testing (pDST) dataset from three sources: the CRyPTIC dataset (Comprehensive Resistance Prediction for Tuberculosis: an International Consortium), a published multi-study compilation, and newly curated literature-derived datasets. The final benchmark contains 54,364 paired WGS-pDST records with broad geographic, lineage, and drug coverage. After harmonizing phenotypes and generating standardized variant features, we evaluated seven models (including classical machine learning and deep learning architectures) across 18 drug-level and six clinical resistance category prediction tasks. Results: XGBoost achieved the highest mean drug-level AUPRC (0.674) and F1-score (0.620) and ranked first in AUPRC for 11 of 18 drugs, whereas WDNN achieved the highest mean AUROC. Random forest yielded the highest mean specificity (0.956) and accuracy (0.933), whereas logistic regression achieved the highest mean recall (0.774), highlighting distinct clinical trade-offs. Drug-level difficulty was highly heterogeneous: rifampicin and isoniazid were predicted robustly, whereas bedaquiline, delamanid, linezolid, and clofazimine remained persistently difficult. In clinical resistance category evaluation, RR-TB, MDR-TB, and pan-susceptibility were well predicted, but XDR-TB and other resistance categories constituted major bottlenecks. Conclusions: Under the largest unified benchmark to date, classical machine-learning methods, particularly XGBoost, provided the strongest precision--recall and F1 performance overall, while neural models remained competitive by AUROC. Emerging drugs (bedaquiline, delamanid, linezolid, clofazimine) and XDR cases remain persistently difficult to predict, identifying key bottlenecks for future method development. This benchmark can serve as a community standard for evaluating MTB resistance prediction and the provided evaluation pipeline offers an actionable baseline for regulatory qualification and clinical decision support system validation, accelerating the translation of WGS-based resistance prediction into practice.