JSTOR

Methods

The study was conducted at the Dr. Josep Trueta Hospital, a 396‐bed teaching hospital in Girona (Catalonia, Spain). In 2002, a proactive program was started to screen 4 categories of high‐risk patients for MRSA: patients transferred from nursing homes or long‐term care facilities, readmitted patients who previously had been infected or colonized with MRSA, patients with more than 30 days of hospitalization in an intensive care unit (ICU) or a surgical ward, and patients exposed to nonisolated, MRSA‐positive patients whose MRSA status was unknown during the exposure period and with whom a room was usually shared until the diagnosis was obtained. Contact isolation precautions, according to Centers for Disease Control and Prevention guidelines,³ and decolonization procedures were implemented when MRSA was detected.

The costs of the program were retrospectively calculated, taking into account 3 main components: the time given by the infection control nurse to the program, the cost of screening and follow‐up cultures, and the cost of isolation and decolonization procedures for MRSA‐positive patients. To analyze the attributable costs due to the incremental length of stay hospital‐acquired MRSA infections in 2002, we used the approach proposed by Wakefield et al.⁴ This approach uses the Appropriateness Evaluation Protocol to determine whether each day of hospitalization is necessary according to 27 criteria. A day of hospitalization was classified as being associated with MRSA infection if at least 1 of the following criteria observed in the medical record was attributable to the infection⁵: diagnostic procedures, therapeutic procedures, and need for nursing services related to the infection. All medical records were reviewed using a 2‐phase approach. First, all MRSA infection attributable days were identified. Second, a new review was conducted excluding all information related to the infection. If no Appropriateness Evaluation Protocol criteria were found, the hospital day was considered attributable solely to the MRSA infection. If a patient was readmitted because of a nosocomial MRSA infection, the total length of stay was considered to be attributable exclusively to the infection. The sum of the MRSA infection–attributable days was calculated for every infected patient.

The cost of a patient‐day was calculated for ICU and non‐ICU hospital wards by means of a multiple distribution method used by the analytic account department of the hospital. This method allocates variable costs (the costs of medication, laboratory tests, imaging studies, and food) and the costs of all other supplies used in the delivery of patient care (housekeeping, repairs, and maintenance) to every hospitalization ward. The average cost per patient‐day is obtained by dividing the total cost per ward and the number of patient‐days in that ward for a specific period. In 2002, the mean cost of 1 hospital‐day was €972 for the ICU and €256 for the conventional wards.

Results

From January to through 2002, a total of 214 patients were screened. The average rate of detection of MRSA colonization was 37% overall (23.1% among patients from long‐term care facilities, 65.8% among readmitted patients with MRSA colonization, 14.3% among patients with more than 30 days of hospitalization in an ICU or in surgical wards, and 13.0% among patients sharing the same room with an MRSA‐positive patient). The estimated annual cost of the program was €10,261 (Table 1). In 2002, there were 37 new cases of nosocomial MRSA infection (Table 2). The patients with pneumonia or respiratory infection had the lowest number of MRSA infection–attributable days, because in most instances the infection was concomitant with receipt of mechanical ventilation and it was not possible to consider these days as attributable solely to the MRSA infection. The economic burden caused by MRSA infections in 1 year was €101,000, and the mean estimated cost of a single infection was €2,730. The cost of the screening program (€10,261) would be justified if it prevented at least 4 MRSA infections (€2,730 per infection × 4 infections = €10,920).

Table 1.  Annual Costs of the Screening Program and of Implementing Isolation and Decolonization Procedures for Patients With Methicillin‐Resistant Staphylococcus aureus Colonization or Infection

Open New Window

Table 2.  Summary of 37 New Cases of Nosocomial Methicillin‐Resistant Staphylococcus aureus Infection in 2002 and Associated Costs

Open New Window

Discussion

Our results show that both patients previously infected or colonized with MRSA who are readmitted and residents of long‐term care facilities and nursing homes can constitute hidden reservoirs of MRSA, so that prompt implementation of contact precautions and decolonization procedures can result in a considerable reduction in the probability of transmission during the hospitalization. To calculate the cost of MRSA infections we chose the method of Wakefield et al.,⁴ because it has been used in studies of the attributable costs of surgical‐site infection⁵ and nosocomial bacteremia⁶ and in a study analyzing the costs of an MRSA screening program.⁷ Since agreement among observers is greater than 75%,⁶ this method may be considered highly reliable. Our study shows that the cost of the screening program would be covered if it only prevented 4 infections per year in our hospital. In extrapolating these results, we may conclude that such programs are beneficial in terms of cost savings provided they reduce the number of nosocomial MRSA infections by a minimum of 11% (4 of 37 MRSA infections is 11%). Our findings are similar to those previously reported from larger hospitals^7‐9 and indicate that policies for screening high‐risk patients can be considered in healthcare facilities where MRSA is a relevant nosocomial pathogen.