Date on Master's Thesis/Doctoral Dissertation

8-2003

Document Type

Master's Thesis

Degree Name

M.A.

Department

Mathematics

Committee Chair

Bradley, Mary Elizabeth

Subject

Drug resistance in microorganisms--Mathematical models; Intensive care units

Abstract

Antibiotic resistance is a problem causing growing concern in the medical community, leading some to speculate that a return to the preantibiotic era is imminent. The problem of antibiotic resistance is particularly significant in the intensive care unit (ICU), due to the weakened immune responses of the patients and quantity of antibiotics administered. One theory proposes that the policy of cycling, or rotating, the antibiotics used in the ICU may minimize the development of resistance. Few clinical trials investigating the effects of cycling have been conducted, and many questions concerning the impact of cycling policies are unanswered at this point. In this thesis, we develop and analyze a mathematical model designed to examine resistance dynamics in the ICU in response to a cycling policy. The uncertainty analysis performed on the model evaluates the variability of the model outcome due to the uncertainty in estimating input values. The chosen method of analysis is Latin Hypercube Sampling (LHS). This uncertainty analysis is extended with the Latin Hypercube Sampling/Partial Rank Correlation (LHS/PRC) sensitivity analysis technique, which identifies the input variables that have the greatest effect on the model outcome. The analysis results show that the prediction imprecision of the model is quite high, leading us to conclude that the model's potential as an investigative tool cannot be fully realized until input values can be estimated with greater certainty.

Download

Share

COinS