Objective: To evaluate SAPS 3 performance in Spain, assessing discrimination and calibration in a multicenter study.
Design: A prospective, multicenter study was carried out.
Patients And Setting: A prospective cohort study was performed in Spanish hospitals between 2006 and 2011.
Measurements And Results: A total of 2171 patients were included in the study. The mean age was 61.4±16.09 years, the ICU mortality was 11.6%, and hospital mortality 16.03%. The SAPS 3 score was 46.29±14.34 points, with a probability of death for our geographical area of 18.57%, and 17.97% for the general equation. The differences between observed-to-predicted mortality were analyzed with the Hosmer-Lemeshow test, which yielded H=31.71 (p<0.05) for our geographical area and H=20.05 (p<0.05) for the general equation. SAPS 3 discrimination with regard to hospital mortality, tested using the area under the ROC curve, was 0.845 (0.821-0.869).
Conclusion: Our study shows good discrimination of the SAPS 3 system in Spain, but also inadequate calibration, with differences between predicted and observed mortality. There are more similarities with regard to the general equation than with respect to our geographical area equation, and in both cases the SAPS 3 system overestimates mortality. According to our results, Spanish ICU mortality is lower than in other hospitals included in the multicenter study that developed the SAPS 3 system, in patients with similar characteristics and severity of illness.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.medin.2013.06.003 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Lipoprotein(a) Atherosclerotic Cardiovascular Disease Risk Score Development and Prediction in Primary Prevention From Real-World Data.
Circ Genom Precis Med
January 2025
Mary and Steve Wen Cardiovascular Division, Department of Medicine, University of California, Los Angeles. (W.F., N.D.W.).
Background: Lp(a; Lipoprotein[a]) is a predictor of atherosclerotic cardiovascular disease (ASCVD); however, there are few algorithms incorporating Lp(a), especially from real-world settings. We developed an electronic health record (EHR)-based risk prediction algorithm including Lp(a).
Methods: Utilizing a large EHR database, we categorized Lp(a) cut points at 25, 50, and 75 mg/dL and constructed 10-year ASCVD risk prediction models incorporating Lp(a), with external validation in a pooled cohort of 4 US prospective studies.
Accurate survival prediction of patients with long-bone metastases is challenging, but important for optimizing treatment. The Skeletal Oncology Research Group (SORG) machine learning algorithm (MLA) has been previously developed and internally validated to predict 90-day and 1-year survival. External validation showed promise in the United States and Taiwan.
View Article and Find Full Text PDFMacrophage heterogeneity and oncogenic mechanisms in lung adenocarcinoma: insights from scRNA-seq analysis and predictive modeling.
Front Immunol
January 2025
Tianjin Chest Hospital, Tianjin University, Tianjin, China.
Background: Macrophages play a dual role in the tumor microenvironment(TME), capable of secreting pro-inflammatory factors to combat tumors while also promoting tumor growth through angiogenesis and immune suppression. This study aims to explore the characteristics of macrophages in lung adenocarcinoma (LUAD) and establish a prognostic model based on macrophage-related genes.
Method: We performed scRNA-seq analysis to investigate macrophage heterogeneity and their potential pseudotime evolutionary processes.
Integrating traditional machine learning with qPCR validation to identify solid drug targets in pancreatic cancer: a 5-gene signature study.
Front Pharmacol
January 2025
Department of General Surgery, The Fourth Affiliated Hospital of Guangxi Medical University, Liuzhou, China.
Background: Pancreatic cancer remains one of the deadliest malignancies, largely due to its late diagnosis and lack of effective therapeutic targets.
Materials And Methods: Using traditional machine learning methods, including random-effects meta-analysis and forward-search optimization, we developed a robust signature validated across 14 publicly available datasets, achieving a summary AUC of 0.99 in training datasets and 0.
Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning.
Gates Open Res
January 2025
University of Virginia, Charlottesville, Virginia, USA.
Background: The TaqMan Array Card (TAC) is an arrayed, high-throughput qPCR platform that can simultaneously detect multiple targets in a single reaction. However, the manual post-run analysis of TAC data is time consuming and subject to interpretation. We sought to automate the post-run analysis of TAC data using machine learning models.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!