bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/457465

Effect of a Sepsis Prediction Algorithm on Patient Mortality, Length of Stay, and Readmission

Burdick, H.·Pino, E.·Gabel-Comeau, D.·McCoy, A.·Gu, C.·Roberts, J.·Slots, J.·Saber, N.·Hoffman, J.·Das, R.
View original source ↗Export metadata ↓

Abstract

ObjectiveTo validate performance of a machine learning algorithm for severe sepsis determination up to 48 hours before onset, and to evaluate the effect of the algorithm on in-hospital mortality, hospital length of stay, and 30-day readmission.\n\nSettingThis cohort study includes a combined retrospective analysis and clinical outcomes evaluation: a dataset containing 510,497 patient encounters from 461 United States health centers for retrospective analysis, and a multiyear, multicenter clinical data set of real-world data containing 75,147 patient encounters from nine hospitals for clinical outcomes evaluation.\n\nParticipantsFor retrospective analysis, 270,438 adult patients with at least one documented measurement of five out of six vital sign measurements were included. For clinical outcomes analysis, 17,758 adult patients who met two or more Systemic Inflammatory Response Syndrome (SIRS) criteria at any point during their stay were included.\n\nResultsAt severe sepsis onset, the MLA demonstrated an AUROC of 0.91 (95% CI 0.90, 0.92), which exceeded those of MEWS (0.71, P<001), SOFA (0.74; P<.001), and SIRS (0.62; P<.001). For severe sepsis prediction 48 hours in advance of onset, the MLA achieved an AUROC of 0.77 (95% CI 0.73, 0.80). For the clinical outcomes study, when using the MLA, hospitals saw an average 39.5% reduction of in-hospital mortality, a 32.3% reduction in hospital length of stay, and a 22.7% reduction in 30-day readmission rate.\n\nConclusionsThe MLA accurately predicts severe sepsis onset up to 48 hours in advance using only readily available vital signs in retrospective validation. Reductions of in-hospital mortality, hospital length of stay, and 30-day readmissions were observed in real-world clinical use of the MLA. Results suggest this system may improve severe sepsis detection and patient outcomes over the use of rules-based sepsis detection systems.\n\nKEY POINTSO_ST_ABSQuestionC_ST_ABSIs a machine learning algorithm capable of accurate severe sepsis prediction, and does its clinical implementation improve patient mortality rates, hospital length of stay, and 30-day readmission rates?\n\nFindingsIn a retrospective analysis that included datasets containing a total of 585,644 patient encounters from 461 hospitals, the machine learning algorithm demonstrated an AUROC of 0.93 at time of severe sepsis onset, which exceeded those of MEWS (0.71), SOFA (0.74), and SIRS (0.62); and an AUROC of 0.77 for severe sepsis prediction 48 hours in advance of onset. In an analysis of real-world data from nine hospitals across 75,147 patient encounters, use of the machine learning algorithm was associated with a 39.5% reduction in in-hospital mortality, a 32.3% reduction in hospital length of stay, and a 22.7% reduction in 30-day readmission rate.\n\nMeaningThe accurate and predictive nature of this algorithm may encourage early recognition of patients trending toward severe sepsis, and therefore improve sepsis related outcomes.\n\nSTRENGTHS AND LIMITATIONS OF THIS STUDYO_LIA retrospective study of machine learning severe sepsis prediction from a dataset with 510,497 patient encounters demonstrates high accuracy up to 48 hours prior to onset.\nC_LIO_LIA multicenter clinical study of real-world data using this machine learning algorithm for severe sepsis alerts achieved reductions of in-hospital mortality, length of stay, and 30-day readmissions.\nC_LIO_LIThe required presence of an ICD-9 code to classify a patient as severely septic in our retrospective analysis potentially limits our ability to accurately classify all patients.\nC_LIO_LIOnly adults in US hospitals were included in this study.\nC_LIO_LIFor the real-world section of the study, we cannot eliminate the possibility that implementation of a sepsis algorithm raised general awareness of sepsis within a hospital, which may lead to higher recognition of septic patients, independent of algorithm performance.\nC_LI

Explore related subjects

bioinformaticsPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Burdick, H., Pino, E., Gabel-Comeau, D., McCoy, A., Gu, C., Roberts, J., Slots, J., Saber, N., Hoffman, J., Das, R.. 2018-10-30. Effect of a Sepsis Prediction Algorithm on Patient Mortality, Length of Stay, and Readmission. https://doi.org/10.1101/457465

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

RoDPAA: A Robust Dissimilarity-Based Principal Angle Analysis Framework for Multi-View Microbiome Data Integration

Multiview microbiome studies profile the same participants across body sites or molecular layers, but integration is hampered by sparse, compositional data, modest sample sizes and methods that recognize only exact sharing. We present RoDPAA (Robust Dissimilarity-based Principal Angle Analysis), a framework that represents each view through any data-appropriate dissimilarity, such as Bray--Curtis, without log-ratio transformation or likelihood assumptions. RoDPAA aligns view-specific signal subspaces through principal angles, captures complete and partial sharing, and tests by permutation how many components are shared, using only eigendecomposition and little tuning. In Gaussian simulations, RoDPAA matched JIVE and MOFA given true ranks, ran faster and alone recovered the shared dimension in every replicate. In two inflammatory bowel disease (IBD) cohorts with sparse, zero-inflated data, Bray--Curtis-based RoDPAA recovered shared structure that was stable to prevalence filtering, whereas log-ratio-based alternatives were not. Its leading microbiome--metabolome factors, learned without disease labels, were the only ones associated with IBD.

bioinformatics↗
bioRxiv2026

Deducing splicing events from transcript sets with splicelogic

Long-read RNA sequencing captures a greater proportion of full-length transcripts, enabling more accurate isoform identification, discovery of novel isoforms, and more precise differential transcript usage analysis. However, comparing transcripts, whether between conditions or between sets such as novel and annotated isoforms, yields complex isoform changes that are hard to manipulate, summarize, and visualize. This is compounded by vertical integration of some tools, where downstream analysis depends on the specific output of a particular upstream tool. To address this gap, we have developed splicelogic, an R/Bioconductor package that makes transcript-set comparisons biologically interpretable by turning them into discrete splicing events, such as exon skipping, alternative splice sites, intron retention, and alternative transcription start and end sites. We designed splicelogic to integrate easily with existing RNA-seq workflows and tools. The splicelogic functions are modular, readable, and extensible, making it easy for users to adapt core logic to their own custom scientific questions or novel splicing definitions. By obtaining interpretable splicing events from transcript changes, splicelogic enables downstream analyses, including annotation of RNA-binding protein recognition sites, alongside summary statistics and visualization. The splicelogic package is publicly available and distributed through Bioconductor.

bioinformatics↗
bioRxiv2026

Contrastive Learning Improves Generalization on M2OR Olfactory Binding Dataset

Olfaction-based biosensors stand to have profound impacts in fields such as medical diagnostics, national security, and drug discovery. A promising mode of vapor phase biodetection mimics volatile organic compound-olfactory receptor (VOC-OR) interactions with target-specific protein or peptide probes; however, lack of experimental coverage remains a core bottleneck. Current computational approaches aim to improve prediction of VOC-OR binding to reduce experimental burden, and current machine learning approaches fail to demonstrate a level of generalization that addresses current engineering challenges. Here, using the Molecular to Olfactory Dataset (M2OR) and Insect Olfactory Binding Protein Database (iOBPdb) datasets, data partitioning methods and their performance are evaluated on a neural network and contrastive learning framework on such VOC-OR binding datasets, varying: model training, architecture, and structure of the input data distribution. It was found that holding out certain chemical or proteomic motifs from the training data diminished model performance with a two layer neural network model. The model was able to generalize to state of the art performance (0.88 AUCROC) on every data split by applying a contrastive learning objective defined by shared molecular constituents and a stochastic sampler. This suggests the contrastive objective provides a strong inductive bias for neural networks to predict novel VOC-OR pairs. These findings highlight that model performance on pairwise biological interaction datasets can be inflated through identity leakage from constituent overlap, and emphasize the need for more robust evaluation protocols. Furthermore, we show contrastive regularization mitigates dataset degeneracy through enforcement of compatibility in the models latent geometry that may extend to other binder target-datasets where identity leakage confounds evaluation.

bioinformatics↗