Identifying Patients at Risk for Postoperative Pneumonia Using a Data Mining Approach

2.50
Hdl Handle:
http://hdl.handle.net/10755/147323
Type:
Presentation
Title:
Identifying Patients at Risk for Postoperative Pneumonia Using a Data Mining Approach
Abstract:
Identifying Patients at Risk for Postoperative Pneumonia Using a Data Mining Approach
Conference Sponsor:Sigma Theta Tau International
Conference Year:2005
Author:Berger, Anne M., PhD, MBA, RN
P.I. Institution Name:Children's Hospital Boston
Title:Director, Nursing Core Metrics
Nurses provide care for surgical patients throughout all phases of their surgical experiences. The development of a postoperative infection can be a significant and potentially life-threatening setback for the surgical patient. Each year in the U.S., approximately 2 million patients develop nosocomial infections, and roughly 88 thousand die from them. Despite advances in knowledge and technology, postoperative pneumonia remains a commonly occurring complication. Over the past decade, the overall rate of nosocomial infections has remained constant, with infections at three major sites; urinary tract infections (UTI), pneumonia, and primary blood stream infections representing the majority of nosocomial infections. Although pneumonia ranks second to UTIs, it has the highest mortality rate (20% to 46%), and it is significantly more costly to treat. Guidelines for the prevention of nosocomial pneumonia in surgical patients remain unresolved because research to date has not fully identified the factors associated with its development. Using Donabedian's Model of Health Care Quality as a framework, the outcome of postoperative pneumonia as it is related to structures and processes of care was examined. Data mining classification methods (Attribute Importance, Support Vector Machine, Naive Bayes and Decision Trees) were used to analyze 122 structure and process variables from the hospital courses of surgical patients who developed postoperative pneumonia and those who did not. This single-site exploratory study revealed factors that contributed significantly to the outcome of postoperative pneumonia in these surgical patients. The development of a predictive model with 72% accuracy in predicting positive pneumonia cases was possible. This knowledge will contribute to the development of evidence based nursing interventions and health promotion strategies aimed at the reduction of risk of postoperative pneumonia in surgical patients.
Repository Posting Date:
26-Oct-2011
Date of Publication:
17-Oct-2011
Sponsors:
Sigma Theta Tau International

Full metadata record

DC FieldValue Language
dc.typePresentationen_GB
dc.titleIdentifying Patients at Risk for Postoperative Pneumonia Using a Data Mining Approachen_GB
dc.identifier.urihttp://hdl.handle.net/10755/147323-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Identifying Patients at Risk for Postoperative Pneumonia Using a Data Mining Approach</td></tr><tr class="item-sponsor"><td class="label">Conference Sponsor:</td><td class="value">Sigma Theta Tau International</td></tr><tr class="item-year"><td class="label">Conference Year:</td><td class="value">2005</td></tr><tr class="item-author"><td class="label">Author:</td><td class="value">Berger, Anne M., PhD, MBA, RN</td></tr><tr class="item-institute"><td class="label">P.I. Institution Name:</td><td class="value">Children's Hospital Boston</td></tr><tr class="item-author-title"><td class="label">Title:</td><td class="value">Director, Nursing Core Metrics</td></tr><tr class="item-email"><td class="label">Email:</td><td class="value">anneberger@earthlink.net</td></tr><tr><td colspan="2" class="item-abstract">Nurses provide care for surgical patients throughout all phases of their surgical experiences. The development of a postoperative infection can be a significant and potentially life-threatening setback for the surgical patient. Each year in the U.S., approximately 2 million patients develop nosocomial infections, and roughly 88 thousand die from them. Despite advances in knowledge and technology, postoperative pneumonia remains a commonly occurring complication. Over the past decade, the overall rate of nosocomial infections has remained constant, with infections at three major sites; urinary tract infections (UTI), pneumonia, and primary blood stream infections representing the majority of nosocomial infections. Although pneumonia ranks second to UTIs, it has the highest mortality rate (20% to 46%), and it is significantly more costly to treat. Guidelines for the prevention of nosocomial pneumonia in surgical patients remain unresolved because research to date has not fully identified the factors associated with its development. Using Donabedian's Model of Health Care Quality as a framework, the outcome of postoperative pneumonia as it is related to structures and processes of care was examined. Data mining classification methods (Attribute Importance, Support Vector Machine, Naive Bayes and Decision Trees) were used to analyze 122 structure and process variables from the hospital courses of surgical patients who developed postoperative pneumonia and those who did not. This single-site exploratory study revealed factors that contributed significantly to the outcome of postoperative pneumonia in these surgical patients. The development of a predictive model with 72% accuracy in predicting positive pneumonia cases was possible. This knowledge will contribute to the development of evidence based nursing interventions and health promotion strategies aimed at the reduction of risk of postoperative pneumonia in surgical patients.</td></tr></table>en_GB
dc.date.available2011-10-26T09:31:15Z-
dc.date.issued2011-10-17en_GB
dc.date.accessioned2011-10-26T09:31:15Z-
dc.description.sponsorshipSigma Theta Tau Internationalen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.