&&ReWrAp:HEADERFOOTER:0:ReWrAp&&

Date of Award

Fall 2011

Degree Type

Thesis

Degree Name

Master of Science (MSc)

Department

Health Research Methodology

Supervisor

Mark Loeb

Co-Supervisor

Stephen Walter

Language

English

Committee Member

Kevin Brazil

Abstract

Antibiotic use is generally regarded as the major driver for resistance. Many studies reporting an association between antibiotic use and the emergence of resistance have been published. However, most studies have significant limitations such as single center data with comparably low number of cases, using retrospective designs with limited data availability, ecological studies with lack of assessing the individual level and risk for ecological fallacy, and inappropriate selection of controls in case-control studies.

A cohort study in adult patients hospitalized in 15 participating acute care hospital sites in Ontario, Canada, was conducted from April 1 2005 to June 30 2006. Antibiotic use on the unit level in defined daily doses (DDD) was only available for 3 sites. In order to assess antibiotic use on both the individual as well as on the unit level as a risk factor for resistance, days of therapy (DOT) could be calculated. However, it was unclear whether this approach would results in similar findings as when using DDD. Thus, the impact of using either DDD or DOT on the risk estimates for resistance was assessed for three antimicrobial-bacteria combinations, i.e. fluoroquinolone use and fluoroquinolone resistance in enterobacteriaceae an in Pseudomonas aeruginosa, and the use of betalactams and resistance to third generation cephalosporins in enterobacteriaceae.

The risk estimates for resistance were very similar for all three antimicrobial-bacteria combinations on acute care units, there were some discrepancies on the unit level on intensive care units, and discrepancies on both levels for step down and rehabilitation units.

In conclusion, the approach to use DOT instead of DDD to measure antibiotic utilization revealed similar results. However, the lack of comprehensive information on patient transfers when calculating DOT may bias the findings on units with frequent patient transfers such as intensive care units and step down and rehabilitation units.

McMaster University Library

Share

COinS