357 papers
Epidemiology is the study of how diseases spread through populations and what factors influence their patterns. Rather than focusing on individual patients, this field examines broader trends to identify outbreaks, track transmission, and guide public health decisions. By analyzing data on infection rates and risk factors, researchers work to prevent future health crises and protect communities worldwide.
On Gist.Science, we process every new preprint in this category directly from medRxiv to make these critical findings instantly accessible. For each study, we provide both a plain-language explanation for general readers and a detailed technical summary for specialists. This dual approach ensures that vital insights into disease dynamics are understood clearly and quickly by everyone who needs them.
Explore the latest research below to see how scientists are currently mapping disease trends and developing strategies to safeguard global health.
UKBAnalytica is a comprehensive, extensible R package designed for the UK Biobank Research Analysis Platform that integrates scalable multi-source disease phenotyping, survival-ready cohort construction, and diverse downstream analytical modules to streamline reproducible epidemiological research.
EpiLink is a novel, threshold-free, simulation-based method that estimates the compatibility of recent transmission between pathogen cases by modeling genetic and temporal uncertainties, offering an interpretable alternative to fixed-distance thresholds and supervised models when labeled transmission data are unavailable.
This study developed and validated a Natural Language Processing pipeline using Bio_ClinicalBERT models to accurately identify brain metastases in radiology reports across three Canadian provinces, enabling scalable population-level surveillance of asynchronous cases that are currently under-captured by cancer registries.
This study demonstrates that a genomic wastewater surveillance framework successfully tracked seasonal influenza A subtypes and detected widespread, dairy-cow-associated H5N1 clade 2.3.4.4b across California during the 2024-2025 flu season, revealing a significant discrepancy between consistent environmental detection and sporadic clinical reporting while confirming the absence of mammalian-adaptive mutations.
This study utilizes proteome-wide Mendelian randomization and colocalization analyses to identify 37 causal plasma proteins in rheumatoid arthritis, revealing shared biological mechanisms with cardiovascular diseases and highlighting both established and emerging druggable targets for therapeutic intervention.
A retrospective study of 41 emergency departments in Germany found that the UEFA European Football Championship 2024 resulted in only minimal, non-significant shifts in attendance patterns, suggesting the event did not place a major strain on emergency healthcare services.
This paper utilizes a mathematical model to demonstrate that traveler screening alone is insufficient for preventing the cross-border spread of Bundibugyo virus, necessitating the implementation of post-arrival quarantine and monitoring to effectively detect infections with long incubation periods.
This study of U.S. women reveals that pregnancy loss is significantly associated with adverse cardiometabolic and behavioral risk factors, including obesity, smoking, and dyslipidemia, suggesting it serves as a key marker of systemic metabolic vulnerability that warrants integration into preconception care strategies.
This paper analyzes a decade of the CDC's FluSight Challenge, revealing that ensemble forecasts and sustained team participation consistently yield the most accurate influenza predictions across diverse model types and seasons, thereby highlighting the value of collaborative forecasting for public health preparedness.
While polygenic risk scores for type 2 diabetes and physical activity-related traits independently predict disease incidence and comorbidities in the FinnGen cohort, incorporating physical activity genotypes into predictive models does not significantly improve accuracy beyond the type 2 diabetes score, whereas adding measured body mass index and smoking status substantially enhances prediction.