Quantifying Cost-Effectiveness of Controlling Nosocomial Spread of Antibiotic-Resistant Bacteria: The Case of MRSA

Background: The costs and benefits of controlling nosocomial spread of antibiotic-resistant bacteria are unknown. Methods: We developed a mathematical algorithm to determine cost-effectiveness of infection control programs and explored the dynamical interactions between different epidemiological variables and cost-effectiveness. The algorithm includes occurrence of nosocomial infections, attributable mortality, costs and efficacy of infection control and how antibiotic-resistant bacteria affect total number of infections: do infections with antibiotic-resistant bacteria replace infections caused by susceptible bacteria (replacement scenario) or occur in addition to them (addition scenario). Methicillin-resistant Staphylococcus aureus (MRSA) bacteremia was used for illustration using observational data on S. aureus bacteremia (SAB) in our hospital (n = 189 between 2001-2004, all being methicillin-susceptible S. aureus [MSSA]). Results: In the replacement scenario, the costs per life year gained range from (sic)45,912 to (sic)6590 for attributable mortality rates ranging from 10% to 50%. Using (sic)20,000 per life year gained as a threshold, completely preventing MRSA would be cost-effective in the replacement scenario if attributable mortality of MRSA is >= 21%. In the addition scenario, infection control would be cost saving along the entire range of estimates for attributable mortality. Conclusions: Cost-effectiveness of controlling antibiotic-resistant bacteria is highly sensitive to the interaction between infections caused by resistant and susceptible bacteria (addition or replacement) and attributable mortality. In our setting, controlling MRSA would be cost saving for the addition scenario but would not be cost-effective in the replacement scenario if attributable mortality would be <21%.