Thesis (Ph. D.)--University of Washington, 2001 Background. Use of intravascular catheters may result in an increased risk of catheter-related bloodstream infection, which is associated with increased morbidity, mortality, and medical costs. Several preventive measures can be used to reduce this risk. Use of antiseptic solutions for skin disinfection at the catheter insertion site is one example. Povidone-iodine and chlorhexidine digluconate are the two most commonly used agents for this purpose. However, the clinical and economic outcomes of these solutions have not been well studied.Methods. The goal of this dissertation is to assess the benefits of chlorhexidine digluconate and povidone-iodine solutions use for the prevention of catheter-related bloodstream infection. This dissertation comprises three separate, yet related projects. First, we evaluate the clinical benefits of both antiseptic solutions using a meta-analytic technique. Second, we determine both clinical and economic benefits of these solutions using decision analytic modeling. Third, we evaluate whether the bootstrap approach as proposed by a group of investigators is an appropriate technique to meta-analyze the risk ratio efficacy.Results. We found that the use of chlorhexidine digluconate for vascular catheter site care decreased the incidence of catheter-related bloodstream infection by 51% compared to povidone-iodine. In addition, using chlorhexidine digluconate for either central or peripheral catheter site care resulted in reduced the incidence of infection, the incidence of death, and total medical cost. We also found that using a bootstrap method as a meta-analytic technique produced invalid results compared to the traditional fixed-effects model.Conclusion. This dissertation suggests that the use of chlorhexidine digluconate for vascular catheter site skin disinfection is a simple and effective means of reducing patient morbidity and mortality, and results in overall cost saving. In addition, the bootstrap method should not be used to meta-analyze the risk ratio outcome, which is the most commonly used outcome measure in meta-analyses and cost-effectiveness analyses.