Related Experiment Videos
Prediction of postoperative nausea and vomiting using a logistic regression model
C C Toner1, C J Broomhead, I H Littlejohn
1Department of Epidemiology and Medical Statistics, London Hospital Medical College, University of London.
British Journal of Anaesthesia
|March 1, 1996
Summary
A validated equation can estimate the risk of postoperative nausea and vomiting (PONV). While not perfect for individual predictions, the model shows good agreement for PONV incidence across risk groups.
Area of Science:
- Anesthesiology
- Clinical Pharmacology
Background:
- Postoperative nausea and vomiting (PONV) is a common adverse event following surgery.
- Predictive models are crucial for identifying patients at high risk of PONV.
Purpose of the Study:
- To validate a previously developed logistic regression equation for predicting PONV risk.
- To assess the equation's accuracy in a diverse patient cohort.
Main Methods:
- Data from 400 patients were collected, including pre-, peri-, and postoperative factors.
- A logistic regression equation incorporating patient factors was used to predict PONV.
- Predicted PONV incidence was compared against actual outcomes.
Main Results:
- The overall incidence of PONV was 36%.
- The equation predicted an overall PONV probability of 27.4%.
- The model correctly classified individual patient outcomes 71% of the time, with good agreement across 16 risk groups.
Conclusions:
- The predictive equation demonstrates good agreement for PONV incidence within defined risk strata.
- Further refinement may be needed for precise individual patient risk stratification.
- The model provides a valuable tool for understanding and managing PONV risk in clinical practice.