Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 4 appraisals
mSphere
Novel machine learning-based approach to identify viral biomarkers of human respiratory emissions from oral and nasal metagenomes
Humans spend approximately 90% of their lives in built environments, making virus transmission indoors a key determinant of health. Environmental sampling of respiratory viral pathogens is often challenging because of frequent non-detect measurements. Non-detect measurements do not differentiate between samples containing low or no pathogens from samples that simply lack respiratory expulsions altogether. This ambiguity can be resolved by scanning samples for a biomarker of human respiratory emissions. To do so, reliable biomarkers for environmental monitoring need to be identified. Ideal biomarkers are prevalent across individuals, abundant, and unique to the human respiratory tract. Here, we present a new machine learning-based approach to query for suitable biomarker candidates from publicly available metagenomes and apply it to identify viral biomarkers of healthy oral and nasal microbiomes. Twelve viral biomarker candidates were selected from 1,232 curated viral operational taxonomic units. The viral biomarker candidates had as much as 63% prevalence across respiratory metagenomes, and prevalence was further increased to 77%-81% by combining two or three biomarkers. Real-time PCR confirmed that these viral biomarkers were prevalent and abundant in nasal swabs and saliva samples. Notably, top candidate biomarkers remained stable and detectable through multiple lab purification steps, increasing confidence in their viral origins and demonstrating their suitability for environmental monitoring. These findings demonstrate that existing metagenomes can be used to identify effective biomarker candidates for environmental sampling.IMPORTANCEDeveloping non-pharmaceutical interventions to reduce virus transmission indoors relies on robust environmental monitoring methods. Monitoring viral pathogens is challenging because of frequent non-detect measurements that introduce uncertainty. For instance, a non-detect measurement could indicate either the absence of the pathogen or simply the lack of human respiratory activity and, thus, exposure. To aid in distinguishing these scenarios, this study identifies viruses from the human respiratory tract using publicly available sequencing data that can be incorporated into environmental monitoring as biomarkers of human respiratory activity. These viral biomarkers will improve indoor monitoring to help enact interventions to mitigate virus transmission. Furthermore, our approach to identify biomarkers from existing metagenomes can be adapted for future biomarker identification in any system.
28 May 2026
Read appraisal →Clinical infectious diseases : an official publication of the Infectious Diseases Society of America
Mycobacterium tuberculosis Transmission in Healthcare Settings: A Systematic Review of Frequency, Risk Factors, and Exposure Conditions
BACKGROUND: Uncertainty regarding factors that govern Mycobacterium tuberculosis transmission in healthcare settings leads to broad and variable exposure definitions and low-yield contact investigations. We systematically reviewed data on M. tuberculosis transmission in healthcare to inform risk stratification and exposure definitions. METHODS: We searched MEDLINE, Embase, CINAHL, and Cochrane databases from inception through December 2024 for studies describing M. tuberculosis transmission from adult patients to healthcare personnel (HCP) or patients. We tabulated data on transmission frequency, circumstances, and risk factors. RESULTS: Of 6695 studies screened, 86 met inclusion criteria, encompassing 1083 source patients, 35 698 exposed individuals, and 2517 secondary cases (7.1% overall transmission risk, 3.8% among studies published since the US Centers for Disease Control and Prevention last updated healthcare ventilation standards in 2005). Transmission was associated with prolonged contact between HCP and source patients (median duration 24 hours, interquartile range [IQR] 24-25 hours) and between patients in shared rooms (median duration 24 hours, IQR 12-24 hours), clinical care without N95 respirator (1232 transmissions/10 149 exposures that included data on face covering use, 12.1%), frequent direct care (185 transmissions/410 exposures, 45.1%), and lack of negative pressure ventilation (1906 transmissions/5670 exposures, 33.6%). Only 8/116 (6.9%) transmissions with documented exposure durations were associated with <8 hours of cumulative exposure and all in poorly ventilated spaces. CONCLUSIONS: Mycobacterium tuberculosis transmission in healthcare is uncommon, particularly in spaces that meet current healthcare ventilation standards. Risk factors for transmission include prolonged exposure, close contact, lack of respirators, and poor ventilation. Defining exposures as >8 hours without a respirator should identify most transmissions and could improve the efficiency of contact investigations.
22 May 2026
Read appraisal →BMJ quality & safety
Retrospective validation study of a large language model approach to screening for intraabdominal surgical site infections for quality and safety reporting
Surgical site infections (SSIs), particularly intra-abdominal (IAB) infections, are challenging to identify and remain a resource-intensive focus of infection prevention programmes. Current automated screening measures rely on discrete data from the electronic health record (EHR), such as microbiology results, diagnosis codes and/or return to the operating room. This approach has poor specificity, and therefore surveillance methods depend heavily on additional manual chart review by trained infection preventionists. Large language models (LLMs) offer an opportunity to improve surveillance by synthesising complex clinical documentation alongside structured data elements.We evaluated the performance of a locally hosted LLM (gpt-35-turbo-16k) to improve IAB SSI screening using perioperative clinical notes and microbiology results. The model analysed documentation across the perioperative period (3 days before through 30 days after surgery) to generate case-level SSI summaries and likelihood assessments. We compared the performance of this tool against the current EHR-based screening workflow.Among 1977 abdominal surgical cases, including 56 with confirmed IAB SSIs, the LLM screened 104 cases as high risk, identifying all infections (negative predictive value (NPV) 100%) and achieving a positive predictive value (PPV) of 53.8%. In contrast, the EHR-based workflow identified 288 cases for further review, with a PPV of only 19.4% and the same NPV of 100%. Analysis of 57 224 notes required ~107 million tokens, translating to approximately USD 0.05 per case.An LLM-based approach to SSI surveillance has the potential to substantially improve efficiency while remaining highly accurate and cost-effective. By reducing reliance on manual chart review, this strategy could allow infection preventionists to shift their attention from surveillance toward quality improvement and patient safety initiatives.
20 May 2026
Read appraisal →PloS one
Artificial intelligence for monitoring hand hygiene compliance in healthcare settings: A scoping review
BACKGROUND: Hand hygiene is a fundamental measure for preventing healthcare-associated infections, yet traditional monitoring methods are significantly limited by the Hawthorne effect, high resource demands, and an inability to assess procedural quality. Artificial intelligence (AI) technology has emerged as a transformative, automated, and objective approach to address these long-standing challenges. OBJECTIVE: This scoping review sought to systematically map the existing evidence, technical pathways, performance metrics, and implementation challenges of AI for monitoring hand hygiene compliance in healthcare settings. METHODS: Following the Joanna Briggs Institute (JBI) methodological framework and PRISMA-ScR guidelines, we searched five major databases (PubMed, Scopus, Embase, Web of Science, and IEEE Xplore) for articles published between January 2000 and September 2025, supplemented by grey literature searching and backward citation tracking. Two reviewers independently screened records, assessed full-text reports for eligibility, and extracted data, which were synthesized using descriptive statistics and thematic analysis. RESULTS: Of 800 records identified through database and supplementary searches, 45 studies (2007-2025) were included. The primary technical pathways identified were computer vision (53.3%), wearable sensors (24.4%), Internet of Things-integrated systems (13.3%), and radar/radio frequency-based systems (8.9%). While computer vision achieved high accuracy (95%) in setting-specific ICU models, performance dropped to 56% in generalizable models. Wearable systems demonstrated portability but showed 5%-10% lower specificity than vision-based approaches. Most evidence is derived from small-scale technical validations, with a significant lack of formal fairness analysis and evaluation of clinical workflows or cost-effectiveness. CONCLUSION: AI-based hand hygiene monitoring shows promise for supporting more objective and scalable hand hygiene surveillance in healthcare settings. However, the field remains at a largely pre-translational stage. Future research should shift from technical feasibility toward implementation science, focusing on establishing standardized motion databases, evaluating ethical governance (e.g., privacy and automation bias), and conducting pragmatic trials to demonstrate sustained clinical benefit and organizational sustainability.
24 Apr 2026
Read appraisal →