Document Type


Date of Degree

Fall 2013

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Perencevich, Eli

Second Advisor

Herwaldt, Loreen

First Committee Member

Schweizer, Marin

Second Committee Member

Diekema, Daniel

Third Committee Member

Dawson, Jeffrey

Fourth Committee Member

Chrischilles, Elizabeth

Fifth Committee Member

Smith, Tara


Staphylococcus aureus causes an assortment of infections that range from mild skin infections to bacteremia or necrotizing pneumonia. Patients with S. aureus infections may suffer poor outcomes such as extended hospital stay and death. The goal of this study was to improve outcomes of patients with S. aureus infections by examining microbial characteristics of S. aureus associated with poor clinical outcomes, and comparative effectiveness of S. aureus treatment options for patients with S. aureus infections. Additionally, methods to prevent S. aureus infections among hospitalized patients were assessed.

We performed a two-hospital retrospective cohort study to identify microbial characteristics, patient characteristics, or antimicrobial treatments that were predictors of mortality or length of stay among patients with methicillin-resistant S. aureus (MRSA) pneumonia. We found increased age (> 54 years) (hazard ratio [HR]: 4.49; 95% confidence interval [CI]: 1.64-12.33), intensive care unit (ICU) admission (HR: 5.25; CI: 1.52-18.21), and having a hospital-onset pneumonia (HR: 0.32; CI: 0.13-0.75) were associated with mortality while admission to the ICU (odds ratio [OR]: 7.34; CI: 3.58-15.04), increased age (> 54 years) (OR: 2.27; CI: 1.19-4.35), having a hospital-onset pneumonia (OR: 3.60; CI: 1.26-10.28), and receiving vancomycin (OR: 10.85; CI: 3.68-32.00) were predictors of increased length of stay. None of the tested microbial characteristics were associated with poor outcomes.

We also completed a multicenter retrospective cohort study to compare the effect of beta-lactams versus vancomycin (both empiric and definitive therapy) on mortality for patients with methicillin-susceptible S. aureus (MSSA) bacteremia who were admitted to Veteran Affairs Medical Centers. We found an increased hazard of mortality for patients who received empiric treatment with a beta-lactam compared with vancomycin (HR: 1.19, 95% CI: 1.00-1.42). However, we observed a protective effect among patients who received definitive treatment with a beta-lactam compared with vancomycin (HR: 0.66; CI: 0.50-0.87).

In 2007, 2009-2011, we administered surveys that focused on the implementation of the Institute for Healthcare Improvement's (IHI) MRSA bundle to reduce hospital-onset MRSA infections to infection preventionsts who worked in Iowa hospitals. By the end of the study period, most hospitals implemented a hand hygiene program (range: 87%-94%), placed infected (range: 97%-100%) or colonized patients (range: 77%-92%) on contact precautions, performed active surveillance culturing to identify colonized patients, and monitored the effectiveness of environmental cleaning (range: 23%-71%; P < 0.001).

To improve patient outcomes, physicians should provide beta-lactams for definitive treatment of patients with MSSA bacteremia. However, the most effective method to improve outcomes is to prevent S. aureus infections from occurring. This study provides benchmark data that infection prevention staff in rural hospitals throughout the U.S. can use to compare their practices with Iowa hospitals.


agr, Bacteremia, MRSA, Pneumonia, PVL, Staphylococcus aureus


x, 168 pages


Includes bibliographical references (pages 151-168).


Copyright 2013 Jennifer Sue McDanel