Clinical Deterioration Prediction

Predictive scoring, early warning, machine learning for clinical deterioration prediction

Predicting Risk of Serious Bacterial Infections in Febrile Children in the Emergency Department

A prospective study from the UK of the ability to predict serious bacterial infections (SBI) via a logistic regression model using clinical and biomarker variables. Investigators used data from 1101 children (median age 2.4 years) who had presented to an ED for fever and who had required laboratory investigation. About a quarter of this group were diagnosed with a SBI, including pneumonia. The diagnostic model discriminated well between pneumonia and no SBI and between other SBIs and no SBI. Model updating yielded good calibration with good performance at both high-risk and low-risk thresholds. Extending the model with procalcitonin and resistin yielded improvements in discrimination.

[su_cite_pediatrics url_fragment = ‘140/2/e20162853’ author = ‘Irwin’ year = ‘2017’] | PubMed 28679639 | Author Search

Multicentre Validation of the Bedside Paediatric Early Warning System Score: A Severity of Illness Score to Detect Evolving Critical Illness in Hospitalised Children

2016-01-15_08-34-33Multicentre (4 hospitals) case-control study to validate the Bedside PEWS score (2,074 patients) in the prediction of cardiopulmonary arrest. The median maximum Bedside PEWS scores for the 12 hours ending 1 hour before the clinical deterioration event were 8 in case patients and 2 in control patients.

Parshuram 2011 (Link) | PubMed 21812993 | Author Search