Dissertation/Thesis Abstract

Patient factors associated with adverse events of hospitalized veterans in infection control isolation
by Grota, Patti Garner, Ph.D., The University of Texas Health Science Center at San Antonio, 2010, 163; 3488185
Abstract (Summary)

This study examined infection control isolation with contact precautions for hospitalized veterans colonized or infected with Methicillin-résistant Staphylococcus aureus (MRSA) and the relationship to adverse events. Study aims were: (1) to describe the characteristics of hospitalized veterans in infection control isolation in a private room (CPI), and infection control isolation in a non private room (CP), (2) to describe the differences in the incidence of adverse events between hospitalized veterans in CPI and in CP, and (3) to predict the probability that a veteran will experience an adverse event in CPI versus CP. An electronic surveillance system was used to select and retrieve computerized medical records from April 2009 through March 2010. Using a retrospective cohort methodology, structured systematic record reviews were conducted by the primary investigator (PI). Patient cultures and room numbers were used to assign the patient record to CPI or CP. Data from 20% of the records were reviewed by trained research assistants with high inter-rater agreement (κ = .97). A physician reviewer validated all adverse events identified by the PI (κ = 1.0).

Of the 316 patient records, 70% (220/316) were in CP and 30% (96/316) were in CPI. The sample was predominately male (97%), white (63%), over 65 years old (44%) and married (49%). There were a total of 104 adverse events identified in 23% (74/316) of the patients. The most frequent adverse events were health care associated infections (20.2%), falls (16.3%), behavioral changes (16.3%), and rapid response interventions (12.5%).

Adverse events were not found to be significantly associated with CPI or CP. Furthermore, there were no differences in the incidence of adverse events in CPI versus CP. Although length of stay (LOS) was positively associated with the occurrence of adverse events (p =.00), the median of 7 days in CP and 9 days in CPI was shorter than 30 days reported in other studies. The lower LOS may have impacted the lack of association between adverse events and CPI/CP. In addition, “close observation” (assignment of a certified nursing assistant to monitor the patient) was identified as a potential confounder and found to be negatively associated with adverse events.

Logistic regression was used to determine the probability of a veteran having an adverse event in CPI and CP. Predictors with bivariate associations with adverse events (p ≤ .01) were entered into the equation: close observation, peripheral vascular disease, dementia, number of unit transfers, fall risk score on admission, LOS, and CPI. Close observation (negatively associated), peripheral vascular disease (negatively associated) and fall risk score (positively associated) had significant partial effects on adverse events, independent of CPI or CP. The model classified 32.6% of veterans who would have an adverse event and 92.6% of veterans who would not have an adverse event accounting for 18.5% of the variance associated with the probability of an adverse event.

The study concludes that isolation type is not a predictor of adverse events, but certain patient characteristics and fall risk scores are predictive of adverse events. Nursing implications include: further research to advance a theoretical framework that can predict factors associated with adverse events, admission assessments that examine patient characteristics and fall risk scores to predict adverse events, and interventions such as the use of close observation to reduce adverse events. Future research should address other factors of the proposed model for isolation and adverse events, replicate this study with other veteran and non-veteran populations related to conditions of isolation, and contribute data that will predict probability of nurse sensitive adverse events, and direct nursing interventions that reduce adverse events in the setting of isolation.

Indexing (document details)
Advisor: Heye, Mary L.
Commitee: Berndt, Andrea E., Larson, Elaine L., Lehman, Cheryl, Patterson, Jan E., Pugh, Mary J.
School: The University of Texas Health Science Center at San Antonio
Department: Nursing
School Location: United States -- Texas
Source: DAI-B 73/04, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Health sciences, Nursing
Keywords: Adverse events, Hospitalization, Infection control isolation, Mrsa, Patient safety, Veterans
Publication Number: 3488185
ISBN: 9781267080318
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest