Use of Longitudinal Surveillance Data to Assess the Effectiveness of Infection Control in Critical Care
A simple method for quantifying nosocomial infection and colonization with multidrug‐resistant organisms is described. This method is applied to the intensive care unit of an academic medical center where longitudinal surveillance data have been used to assess the impact of infection control interventions and antibiotic use.
Received March 13, 2009; accepted May 26, 2009; electronically published October 2, 2009.
Infections caused by multidrug‐resistant organisms are responsible for increases in the length of hospital stay, attributable mortality, and cost of medical care.1 In the intensive care unit (ICU), infections with multidrug‐resistant organisms increase the probability that patients will receive inadequate empiric antimicrobial therapy, which may lead to adverse patient outcomes.2 Data from the National Nosocomial Infectious Surveillance system (now known as the National Healthcare Safety Network) have suggested that there has been an increasing incidence of infection with multidrug‐resistant organisms in the ICUs of US hospitals.3 The increasing rates of infection with multidrug‐resistant organisms in ICUs and the pressure being applied to increase public reporting are driving hospitals to develop strategies to reduce the occurrence of these infections.1,4 For ICUs, these infection control efforts have included the use of hand hygiene, contact isolation, environmental disinfection, active surveillance, antimicrobial stewardship, checklists, and bundles.5 However, what remains unclear is whether these infection control efforts will provide long‐term control of infections with multidrug‐resistant organisms in ICUs. The present article describes the use of longitudinal surveillance data to assess the impact of infection control efforts in critical care.
Methods
Setting. Fletcher Allen Health Care is a 562‐bed academic medical center affiliated with the University of Vermont College of Medicine in Burlington, Vermont. There are 26 beds in the surgical ICU (SICU) and 22 beds in the medical ICU (MICU), and each unit has a 5‐bed open ward. The SICU also includes a 4‐bed pediatric ICU. There are 4 infection control practitioners and an active antibiotic control program managed by an infectious disease specialist and an infectious disease pharmacist. For the antibiotic control program, there has been a prior‐approval process in place and an infectious disease pharmacist employed since the 1980s. Antibiotic restriction was lifted from the ICUs in July 2004.
Surveillance method. The resistance index was developed as a tool for quantifying nosocomial infection and colonization with organisms of epidemiological importance. The resistance index is a rate that is calculated monthly; the numerator is the number of nosocomial isolates (or, in the case of Clostridium difficile, the number of toxin‐positive specimens) of 6 different organisms: (1) methicillin‐resistant Staphylococcus aureus (MRSA), (2) vancomycin‐resistant Enterococcus, (3) C. difficile, (4) fluoroquinolone‐resistant Pseudomonas aeruginosa, (5) ceftazidime‐resistant gram‐negative bacilli, and (6) Stenotrophomonas maltophilia. Only isolates that are recovered more than 48 hours after hospital admission are included. Isolates recovered from patients with cystic fibrosis and during active surveillance for MRSA are excluded. A single patient may be infected or colonized with more than one type of organism, but each type is counted only once per patient. Cultures reflecting both colonization and infection are included. The resistance index is calculated for the 2 ICUs and 6 medical‐surgical wards. A weekly report from the microbiology laboratory is screened, and data on the isolates are entered into a database using Microsoft Access 1997. The denominator is the number of patient‐days for each nursing unit and for the hospital. Organisms are assigned to a specific nursing unit if the patient had resided in that unit for at least 48 hours prior to the positive culture or toxin assay result; the assignments are confirmed by an infection control practitioner. The resistance index database required 8–12 hours of maintenance per month.
Measuring antibiotic use. Antibiotic use was measured in terms of defined daily doses, as proposed by the World Health Organization.6 The following antibiotics were included in our study: fluoroquinolones, carbapenems, ceftriaxone, ceftazidime, aminoglycosides, vancomycin, and piperacillin‐tazobactam; cefepime was not included because of its infrequent use during the study period.
Interventions. In July 2003, the MICU implemented quality‐improvement initiatives that targeted cases of bloodstream infection and cases of ventilator‐associated pneumonia; these initiatives included the use of central line carts, the use of chlorhexidine gluconate–impregnated central venous catheters, and the use of 2% chlorhexidine gluconate and 70% isopropyl alcohol (ChloraPrep; Medi‐Flex) during catheter insertion. These initiatives were gradually adopted by the SICU as well. During the period from February through December 2004, ICU nurses and physicians received additional education and were observed for hand hygiene performance. Active surveillance for MRSA in the ICU began in March 2005. In June 2006, the SICU and the MICU implemented the US Institute for Healthcare Improvement's central line–associated bloodstream infection and ventilator‐associated pneumonia prevention “bundles.” Administrative and medical leaders were engaged and served as mentors and advocates. Environmental cleaning protocols and staff education were enhanced.
Statistical analysis was performed using Stata, version 10.0 (StataCorp). The resistance index and the rate of antimicrobial use were compared using the Poisson distribution. Two‐sided P values of less than .05 were considered to be statistically significant.
Results
The Table presents data on the rates of infection or colonization with multidrug‐resistant organisms (cases per 1,000 patient‐days) and the rates of antimicrobial use (defined daily doses per 1,000 patient‐days) for the SICU and the MICU during 2 time periods: April 2000–June 2004 (the preintervention period) and July 2004–September 2008 (the postintervention period). In the SICU, the rate of C. difficile infection decreased from 2.2 cases per 1,000 patient‐days during the preintervention period to 1.1 cases per 1,000 patient‐days during the postintervention period (ie, by nearly 50%; 
). In the MICU, the rate of C. difficile infection decreased from 2.7 cases per 1,000 patient‐days during the preintervention period to 1.4 cases per 1,000 patient‐days during the postintervention period (ie, by nearly 50%; 
). In the MICU, the rate of recovery of ceftazidime‐resistant gram‐negative bacilli decreased from 2.0 isolates per 1,000 patient‐days during the preintervention period to 1.1 isolates per 1,000 patient‐days during the postintervention period (ie, by nearly 50%; 
). However, in the MICU, the rate of recovery of fluoroquinolone‐resistant P. aeruginosa increased from 1.4 isolates per 1,000 patient‐days during the preintervention period to 2.2 isolates per 1,000 patient‐days during the postintervention period (
). The overall resistance index decreased in both units during the postintervention period.
A closer look at the data indicates certain trends over the 8‐year study period (which was divided into 4 periods in the Figure). During the preintervention period, the resistance index was increasing in both units. For the SICU, the increase in the resistance index during the preintervention period was significant (from 6.6 during the first period to 9.5 during the second period; 
). This trend was reversed during the postintervention period (from 8.9 during the third period to 5.7 during the fourth period; 
). For the MICU, the resistance index also increased during the preintervention period (from 8.0 during the first period to 10.2 during the second period; 
). Similar to what was observed in the SICU, the upward trend in the resistance index was reversed during the postintervention period (from 8.8 during the third period to 7.7 during the fourth period; 
).
Figure. Resistance index (95% confidence interval), by quarter (April 2000–September 2008), in the surgical intensive care unit (A) and medical intensive care unit (B). Each graph is divided into a pre‐ and postintervention period (solid vertical line). The data are further divided into 4 periods of approximately 2 years each, labeled 1–4 (dotted vertical lines), with corresponding resistance indices (no. of isolates per 1,000 patient‐days) and 95% confidence intervals. The interventions are outlined in the Methods section.
The resistance indices were generally stable in the 6 medical‐surgical wards during the 8‐year study period. The overall rate of antimicrobial use in the SICU was higher during the postintervention period than during the preintervention period (366 vs 352 defined daily doses per 1,000 patient‐days; 
), and the overall rate of antimicrobial use in the MICU was higher during the postintervention period than during the preintervention period (603 vs 436 defined daily doses per 1,000 patient‐days; 
).
Discussion
Hospitals should develop a system that can identify longitudinal trends in the detection of multidrug‐resistant organisms.1,4,5 A recent report7 has demonstrated the value of surveillance for central line–associated bloodstream infection due to MRSA in critical care units. The data can provide important information about the impact of infection control activities and antibiotic use. Our study describes a simple surveillance method that provides 8 years of data from critical care units in an academic medical center. This same analysis can be applied to any hospital ward to direct and focus infection control activities. The data that were obtained from this surveillance suggest that infection control initiatives successfully reversed an upward trend in the detection of 6 multidrug‐resistant organisms included in the resistance index in critical care, despite increasing antibiotic use. The overall increase in antibiotic use during the postintervention period in the SICU and MICU could be related to the elimination of the need for prior approval of restricted antibiotics or to the incorporation of newer treatment guidelines for community‐acquired and ventilator‐associated pneumonia.8,9
There were several limitations to our approach. First, our method utilized passive data collection and did not differentiate between infection and colonization. Second, the resistance index was limited to the organisms that were chosen for inclusion. Third, the assignment of cultures to hospital wards may have been compromised by community acquisition and by interhospital and intrahospital transfers of patients. Fourth, genotyping was not performed. Finally, we were unable to determine which infection control strategy was most effective, because multiple interventions were performed simultaneously.
In summary, this longitudinal surveillance strategy was able to capture the impact of infection control interventions in critical care. It also allowed for unit comparisons within hospitals, which may facilitate opportunities for feedback and education.
Acknowledgments
Potential conflicts of interest. Both authors report no conflicts of interest relevant to this article.
References
- 1. Shlaes DM, Gerding DN, John JF Jr, et al. Society for Healthcare Epidemiology of America and Infectious Diseases Society of America Joint Committee on the Prevention of Antimicrobial Resistance: guidelines for the prevention of antimicrobial resistance in hospitals. Infect Control Hosp Epidemiol 1997; 18:275–291.
- 2. Kollef MH. Inadequate antimicrobial treatment: an important determinant of outcome for hospitalized patients. Clin Infect Dis 2000; 31(Suppl 4):S131–S138.
- 3. Siegel JD, Rhinehart E, Jackson M, et al. Management of multidrug‐resistant organisms in healthcare settings, 2006. Am J Infect Control 2007; 35:S165–S193.
- 4. Goldmann DA, Weinstein RA, Wenzel RP, et al. Strategies to prevent and control the emergence and spread of antimicrobial‐resistant microorganisms in hospitals. A challenge to hospital leadership. JAMA 1996; 275:234–240.
- 5. Kollef M, Fraser VJ. Antibiotic resistance in the intensive care unit. Ann Intern Med 2001; 134:298–313.
- 6. WHO Collaborating Centre for Drug Statistics Methodology. About the ATC/DDD system. Available at: http://www.whocc.no/atcddd/. Accessed January 28, 2009.
- 7. Burton DC, Edwards JR, Horan TC, et al. Methicillin‐resistant Staphylococcus aureus central line–associated bloodstream infections in US intensive care units, 1997–2007. JAMA 2009; 301:727–736.
- 8. Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community‐acquired pneumonia in adults. Clin Infect Dis 2007; 44(Suppl 2):S27–S72.
- 9. American Thoracic Society and Infectious Diseases Society of America. Guidelines for the management of adults with hospital‐acquired, ventilator‐associated, and healthcare‐associated pneumonia. Am J Respir Crit Care Med 2005; 171:388–416.