Prediction of postoperative nausea and vomiting using a logistic regression model
| Autor: | L. Strunin, J. G. Powney, I. H. Littlejohn, S. J. W. Evans, C. C. Toner, G. S. Samra, M. G. A. Palazzo, C. J. Broomhead |
|---|---|
| Rok vydání: | 1996 |
| Předmět: |
Male
Vomiting Nausea Logistic regression Postoperative Complications Sex Factors Predictive Value of Tests Risk Factors medicine Humans Anesthesia business.industry Incidence (epidemiology) Regression analysis medicine.disease Analgesics Opioid Logistic Models Anesthesiology and Pain Medicine Motion sickness Predictive value of tests Female medicine.symptom business Postoperative nausea and vomiting |
| Zdroj: | British Journal of Anaesthesia. 76:347-351 |
| ISSN: | 0007-0912 |
| DOI: | 10.1093/bja/76.3.347 |
| Popis: | In a previous study, logistic regression analysis was used to determine the association of independent fixed patient factors with the incidence of postoperative nausea and vomiting (PONV). Female sex, previous history of PONV, use of postoperative opioids, previous history of motion sickness and an interaction between male sex and previous history of PONV were combined in an equation from which risk of PONV could be estimated. The present study was designed to test this equation in a group of patients with wide selection criteria. Data on 400 patients were collected in relation to pre-, per- and postoperative factors which may influence the incidence of PONV. The equation was used to predict PONV, and actual outcome was compared with that predicted. The overall incidence of PONV was 36%. The equation predicted an overall probability of PONV of 27.4%. If the model was used to define individual patients as predicted to have or not to have PONV, it was correct only 71% of the time. However, there was good agreement between the actual incidences of PONV and those predicted among the 16 risk groups created by the model. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|