The Cost of Antimicrobial Resistance in Patients with Nosocomial Staphylococcus Aureus Bactermia
Suzanne T. Phillips
Advisor: Spencer Harpe PharmD, PhD, MPH
Preceptor: Hongjie Liu, PhD, MS
Background: The proportion of Staphylococcus infections caused by methicillin-resistant Staphylococcus-aureus (MRSA) has increased from 22% to 63% in 1995 and 2004 respectively. Blood stream infections, more commonly referred to as bacteremias, represented the majority (75.5%) of hospital-onset MRSA cases. The economic impact of Staphylococcus aureus bacteremia (SAB) merits investigation.
Objective: The purpose of this investigation was to quantify the additional hospitalization charge attributable to methicillin-resistant vs. methicillin-susceptible SABs.
Methods: This was a retrospective cohort analysis within Virginia Commonwealth University’s Health System (VCUHS). Eligible patients were those who had SAB and were discharged from the VCUHS between May 1, 2006 and April 31, 2007. Inclusion criteria include: age > 18 years old and onset of infection > 48 hours post admission. Frequency matching on pre-infection length of stay was used. Patients with SAB were assessed to be either resistant or susceptible to methicillin. Proportions between the groups and continuous variables were evaluated using Pearson’s chi-squared and the Wilcoxon Signed Rank test respectively. The median crude hospital charge for each group was approximated from the data. The median adjusted hospital charge was calculated using a generalized linear model with a gamma distribution and a log link.
Results: During the study period, 219 episodes of SAB were identified. One hundred and thirty-seven patients failed to meet the inclusion criteria. Of the remaining 82 patients, 72 patients met the frequency matching criterion (MRSA = 45, MSSA = 27). The overall median total hospital charge was $140,396 (MRSA = $166,901, MSSA = $86,130.) The crude difference in median charge was $80,771. The final multivariate model included organism, DRG weight, intensive care unit status (ICU), age, discharge status, an interaction between DRG weight and ICU and an interaction between ICU and age. According to the model, a resistant SAB increases total hospital charges by 28%. This model predicted an overall charge of $163,811 (IQR $117,029 to $230,954). The MRSA and MSSA median charges are $186,559 and $145,345 respectively. The difference between the median predicted charges was $41,214.
Conclusions: The 28% increase in hospital charge for resistant SABs is consistent with previous single center investigations. This and previous investigations are limited by their small sample size. A large multi-center investigation needs to be performed. Once the magnitude of economic impact for resistant vs. susceptible SAB becomes clearer, public health initiatives can be addressed.
