Daily Sepsis Research Analysis

BACKGROUND: Delirium complicates up to one in four hospitalisations among older adults, but its clinical sequelae beyond cognitive impairment, functional decline, and mortality are not well characterised. Whether delirium signals broader multisystem vulnerability leading to later adverse outcomes remains uncertain. We aimed to examine the association of in-hospital delirium with the occurrence of a range of adverse outcomes during later hospitalisations in the UK. METHODS: We performed a matched cohort study among hospitalised UK Biobank participants recruited in England, Scotland, and Wales in 2006-10 with linked hospital inpatient records collected from Jan 1, 1997 to Oct 31, 2022. We matched 14 909 individuals with delirium (1:1) to hospitalised control individuals without delirium by age, sex, Hospital Frailty Risk Score, primary diagnosis, episode length of stay, and intensive care unit length of stay of the index episode. The primary outcome was the risk of adverse clinical outcomes following delirium. Delirium was identified by ICD-10 codes, with exposure intensity defined as the number of episodes within 12 months of the index admission. Adverse clinical outcomes were selected based on existing literature and included incident events occurring during subsequent hospitalisations: falls, fractures (any and hip), pressure injury, urinary and faecal incontinence, myocardial infarction, heart failure, stroke, venous thromboembolism, pulmonary embolism, acute kidney injury, gastrointestinal bleeding, pneumonia, and sepsis. Fine-Gray subdistribution hazard models accounted for death as a competing risk and adjusted for socioeconomic covariates. FINDINGS: Over a maximum follow-up of 26 years, the median time to event was 1·2 years (IQR 0·2-3·3) in the delirium group and 1·3 years (0·3-3·6) in the control group. Delirium was associated with a higher risk of 12 of the 15 adverse clinical outcomes than no delirium, which were urinary incontinence (subdistribution hazard ratio 2·01 [95% CI 1·78-2·28]), falls (1·96 [1·78-2·17]), pressure injury (1·72 [1·53-1·93]), acute kidney injury (1·71 [1·57-1·86]), sepsis (1·67 [1·52-1·84]), hip fracture (1·66 [1·39-2·00]), stroke (1·62 [1·41-1·87]), overall fractures (1·56 [1·40-1·74]), pneumonia (1·53 [1·41-1·65]), faecal incontinence (1·53 [1·31-1·78]), heart failure (1·31 [1·18-1·45]), and gastrointestinal bleeding (1·23 [1·09-1·38]). Each additional episode was associated with a 6-17% higher risk. Findings were consistent across sensitivity analyses excluding individuals with short follow-up, those with prevalent dementia, and the least well-matched pairs. INTERPRETATION: In-hospital delirium was consistently and dose-responsively associated with a range of adverse outcomes, independent of frailty and pre-existing dementia, supporting its recognition as a sentinel event indicating longer-term vulnerability. FUNDING: Sigrid Jusélius Foundation, the Osk Huttunen Foundation, the Biomedicum Helsinki Foundation, and Finska Läkaresällskapet.

IMPORTANCE: Prompt antimicrobial therapy is essential in sepsis, but accelerating antimicrobial administration may increase overtreatment. OBJECTIVES: To examine the extent of and factors associated with physician variation in time from emergency department (ED) presentation to antimicrobial administration (hereinafter termed door-to-antimicrobial time) for sepsis and to assess whether faster practice patterns are associated with overtreatment. DESIGN, SETTING, AND PARTICIPANTS: This explanatory mixed-methods study linked a quantitative retrospective cohort (July 1, 2013, to January 31, 2017) involving 30-day patient follow-up with prospective qualitative physician interview data (May 17, 2022, to June 28, 2023) at 4 Utah EDs. Participants included ED attending physicians and their patients meeting sepsis criteria (including intravenous antimicrobial administration) before ED departure. Data analysis occurred from 2021 to 2025. MAIN OUTCOMES AND MEASURES: Assessment for physician door-to-antimicrobial time variation used a likelihood ratio test comparing a linear mixed-effects model incorporating physician-level random intercepts and patient-level covariates with a model without physician random effects. Empirical best linear unbiased predictions of the physician random intercepts (termed physician-predicted mean door-to-antimicrobial times) quantified variation. The primary analysis used a joint mixed-effects shared parameter model to evaluate the association of physicians' door-to-antimicrobial practice patterns with their overtreatment rate (infection ruled out on final retrospective adjudication). Qualitative analysis of semistructured cognitive task analysis interviews compared ED physicians in the fastest and slowest door-to-antimicrobial time quartiles. RESULTS: Quantitative analyses included 88 ED physicians (71 [80.7%] male; median age, 39 [IQR, 35-49] years) and 9810 patients with sepsis (median age, 63 [IQR, 48-75] years), of whom 4635 (50.5%) were female and 3540 (38.6%) received antimicrobials more than 3 hours after ED arrival. The median number of patient encounters per physician was 105 (IQR, 75-129). Physicians' door-to-antimicrobial time varied significantly (likehood ratio test P < .001), with average physician-level estimated mean door-to-antimicrobial time of 184 (95% estimation interval, 146-222) minutes for a typical patient, but was not associated with overtreatment (adjusted odds ratio, 0.98 [95% CI 0.94-1.02] per 10-minute increase in physician estimated mean door-to-antimicrobial time; P = .37). Among 18 physicians interviewed, physicians with faster door-to-antimicrobial times emphasized proactive, parallel task execution and care team coordination, while physicians with slower times described a more reactive and stepwise sepsis evaluation and treatment process. CONCLUSIONS AND RELEVANCE: In this mixed-methods study, ED physicians' antimicrobial administration time for sepsis varied significantly, but faster antimicrobial initiation practice patterns were not associated with overtreatment. Physicians with shorter door-to-antimicrobial times described a proactive, parallel processing approach to sepsis care.

Sepsis-associated encephalopathy (SAE) is common in the intensive care unit (ICU) and portends worse short- and long-term outcomes. To enable real-time bedside use and multicenter deployment, we aimed to develop a parsimonious, transparent day-1 prediction model using routinely available variables while preserving discrimination, calibration, and clinical utility. Using MIMIC-IV (2008-2022), we conducted a single-center retrospective study of adult sepsis patients with KDIGO-defined AKI. Predictors were restricted to the first 24 hours after ICU admission; the endpoint was any in-ICU SAE ("ever" vs "never"). After multiple imputation (m = 5), 44 baseline variables were standardized and entered into LASSO with 20-fold cross-validation. A 3-rule clinical screen (24 hours availability; non-treatment; low collinearity) distilled LASSO-selected features to a four-predictor logistic model; performance was internally validated (bootstrap) and compared with an XGBoost benchmark. SHAP analyses supported interpretability. Among 6780 ICU stays (training n = 4746; validation n = 2034), SAE occurred in 69.8%. The final 4 predictors were age, SAPS II, serum sodium, and mean arterial pressure (MAP). Discrimination was stable (AUC 0.734 training; 0.739 validation) with excellent calibration (validation CITL = -0.045; slope = 0.996; Brier = 0.182). Decision-curve analysis showed greater net benefit than XGBoost across thresholds 0.15 to 0.55; although AUCs were similar, XGBoost calibrated worse (CITL = -0.289; slope = 0.729). SHAP ranked contributions as SAPS II, sodium, age, and MAP, indicating a near-linear sodium-risk rise within 138 to 144 mmol/L, age-related risk above ~70 years, and a U-shaped MAP effect with protection around 55 to 75 mm Hg. We developed and validated a four-factor nomogram that uses only routine day-1 data to stratify SAE risk rapidly and transparently, outperforming a complex learner in calibration and net benefit. This parsimonious, interpretable tool highlights modifiable targets (sodium, individualized MAP) and provides a pragmatic foundation for multicenter validation and EMR-embedded early warning and intervention strategies.