Improving Methodology to Identify True Community Carbapenem-resistant Enterobacteriaceae (CRE) and Assessing Prior Healthcare Exposures to Quantify Risk for CRE Diagnosis Upon Hospital Admission Open Access
Gretzinger, Siyeh (Spring 2020)
Abstract
Antibiotic resistant bacteria are a major threat to public health due to the high morbidity and mortality. Carbapenem-resistant Enterobacteriaceae (CRE) has been identified as an urgent threat given its difficulty to treat. CRE typically presents asymptomatically, making it challenging to determine patients’ colonization status upon hospital admission and whether these infections are arising from the community or from prior healthcare exposures. The purpose of this study was to use information available in the Georgia hospital discharge database to evaluate the misclassification of community-associated CRE and develop a model to predict patient risk of having CRE carriage or infection upon hospital admission. A case-control study was performed using hospital encounter information for 281 cases and 233,786 matched controls obtained from two state-based databases. A set of prior healthcare exposures were evaluated as predictors. Multivariate conditional logistic regression was employed for model development. Odds ratios and respective p-values were calculated to determine direction and strength for each predictor. Model performance was evaluated using ROC and AUC values. Six percent of community CRE was identified as having a missed prior hospitalization thereby warranting re-classification. The final model identified the following variables to be associated with an elevated risk of CRE upon admission: current admission to long-term acute care hospital (OR=17.7, 95% CI:1.40 – 223.29), use of federal health insurance (OR=2.22, 95% CI:1.40 – 3.54), hospital admission with an infection diagnosis in prior year (OR=2.02, 95% CI:1.42 – 2.87), number of short-term acute care hospitalizations (STACH) in prior year (OR=1.18, 95% CI:1.09 – 1.27), age (OR=1.03, 95% CI:1.02 – 1.04), and mean prior STACH length of stay (OR=1.02, 95% CI:1.02 – 1.03). The model had good discriminatory performance with an AUC = 0.76. Data from this study demonstrated that state-wide hospital discharge databases are an important tool that can be used to validate prior healthcare exposures and develop prediction rules to detect patients at higher risk for CRE. Such prediction rules have significant implications for hospitals across the country as they enhance active surveillance methods and support preemptively screening for high-risk patients.
Table of Contents
Chapter I: Background........................................................................................1
Chapter II: Manuscript Abstract.........................................................................7
Introduction.......................................................................................................8
Methods.............................................................................................................10
Results...............................................................................................................14
Discussion.........................................................................................................18
References.........................................................................................................23
Tables................................................................................................................28
Figures...............................................................................................................38
Chapter III: Public Health Implications............................................................39
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Keyword | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Improving Methodology to Identify True Community Carbapenem-resistant Enterobacteriaceae (CRE) and Assessing Prior Healthcare Exposures to Quantify Risk for CRE Diagnosis Upon Hospital Admission () | 2020-04-29 08:52:14 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|