Impact of Contact and Droplet Precautions on the Incidence of Hospital‐Acquired Methicillin‐Resistant Staphylococcus aureus Infection
Objective. To evaluate the efficacy of contact and droplet precautions in reducing the incidence of hospital‐acquired methicillin‐resistant Staphylococcus aureus (MRSA) infections.
Design. Before‐after study.
Setting. A 439‐bed, university‐affiliated community hospital.
Methods. To identify inpatients infected or colonized with MRSA, we conducted surveillance of S. aureus isolates recovered from clinical culture and processed by the hospital's clinical microbiology laboratory. We then reviewed patient records for all individuals from whom MRSA was recovered. The rates of hospital‐acquired MRSA infection were tabulated for each area where patients received nursing care. After a baseline period, contact and droplet precautions were implemented in all intensive care units (ICUs). Reductions in the incidence of hospital‐acquired MRSA infection in ICUs led to the implementation of contact precautions in non‐ICU patient care areas (hereafter, “non‐ICU areas”), as well. Droplet precautions were discontinued. An analysis comparing the rates of hospital‐acquired MRSA infection during different intervention periods was performed.
Results. The combined baseline rate of hospital‐acquired MRSA infection was 10.0 infections per 1,000 patient‐days in the medical ICU (MICU) and surgical ICU (SICU) and 0.7 infections per 1,000 patient‐days in other ICUs. Following the implementation of contact and droplet precautions, combined rates of hospital‐acquired MRSA infection in the MICU and SICU decreased to 4.3 infections per 1,000 patient‐days (95% confidence interval [CI], 0.17‐0.97; 
). There was no significant change in hospital‐acquired MRSA infection rates in other ICUs. After the discontinuation of droplet precautions, the combined rate in the MICU and SICU decreased further to 2.5 infections per 1,000 patient‐days. This finding was not significant (
). In the non‐ICU areas that had a high incidence of hospital‐acquired MRSA infection, the rate prior to implementation of contact precautions was 1.3 infections per 1,000 patient‐days. After the implementation of contact precautions, the rate in these areas decreased to 0.9 infections per 1,000 patient‐days (95% CI, 0.47‐0.94; 
).
Conclusion. The implementation of contact precautions significantly decreased the rate of hospital‐acquired MRSA infection, and discontinuation of droplet precautions in the ICUs led to a further reduction. Additional studies evaluating specific infection control strategies are needed.
Received March 16, 2007; accepted June 14, 2007; electronically published September 18, 2007.
Methicillin‐resistant Staphylococcus aureus (MRSA) continues to be a source of serious hospital‐acquired infections and accounts for more than 35% of such infections.1 Therapeutic options for treating patients infected with resistant strains of S. aureus are limited, and infected patients have an increased length of hospitalization and increased mortality (as high as 40% in some reports).2 Strategies to prevent the transmission of multidrug‐resistant organisms in healthcare settings have been made a priority by the Centers for Disease Control and Prevention (CDC).3
Extensive programs to reduce the spread of hospital‐acquired MRSA infection have been implemented and have shown success. The programs with the greatest success rates were conducted in either small European countries or during short‐term outbreaks.4‐6 They involved the implementation of intensive surveillance for colonization and patient‐isolation strategies, an approach that is not practical over the long term in most healthcare facilities in the United States. Early studies that demonstrated success with active surveillance and decolonization over longer periods were conducted in situations in which the incidence of MRSA infection was low.7 There are insufficient data on the relative impact of specific strategies for controlling the incidence of hospital‐acquired MRSA infection in hospitals,8,9 and there is little consensus regarding appropriate interventions. The CDC does not recommend routine active surveillance as a strategy for the prevention of nosocomial transmission in situations that do not involve an outbreak.10
We conducted a prospective study in our 439‐bed, university‐affiliated community hospital to compare the impact of contact precautions alone with the impact of contact precautions plus droplet precautions on the incidence of hospital‐acquired MRSA infection in intensive care units (ICUs) and non‐ICU patient‐care areas (hereafter, non‐ICU areas).
Methods
Investigation in ICUs
Baseline data on all S. aureus isolates recovered from clinical culture and processed by our clinical microbiology laboratory was obtained from March 1 through June 30, 2002. We reviewed patient records for the individuals from whom MRSA isolates were recovered. The presence of MRSA infection was defined according to CDC guidelines.11 An infection was classified as hospital acquired if the first MRSA isolate was recovered from a clinical sample obtained more than 48 hours after admission. Patients from whom MRSA had been recovered during the 2‐year period previous to the present hospitalization were excluded from the study. The rate of hospital‐acquired MRSA infection was defined as the number of hospital‐acquired infections per 1,000 patient‐days. Monthly rates were calculated for the medical, surgical, cardiac, pediatric, and neonatal ICUs, as well as the cardiovascular recovery unit and the cardiovascular step‐down unit (for the purposes of this study, the latter 2 units were considered ICUs). With the exception of the medical ICU (MICU) and surgical ICU (SICU), all of the ICUs contained private or semiprivate rooms. The SICU has 9 beds, 6 of which are in an open ward and 3 of which are enclosed as single‐bed rooms. The MICU has 12 beds, 9 of which are in an open ward and 3 of which are enclosed as single‐bed rooms.
During the baseline period, (March 1 through June 30, 2002), standard precautions were employed in all ICUs. Standard precautions complied with those recommended by the CDC12 and involve the use of gloves whenever a healthcare worker anticipates contact with a patient’s blood or body fluids or with any contaminated item. Gloves are to be removed and hands washed immediately with soap and water or an alcohol‐based hand cleanser before touching any noncontaminated surface or moving on to another patient. The healthcare worker is to wear a gown, surgical mask, and eye shield if there is a likelihood of splashes of blood or body fluid.
The first intervention period in the ICUs began on July 1, 2002, and ended on April 30, 2004. Contact precautions were implemented for all patients who had MRSA recovered from a clinical culture of samples obtained from any anatomical site. If the clinical isolate was obtained from a respiratory tract sample, the patient was also placed under droplet precautions. Contact precautions require that healthcare workers put on gloves upon entering the patient's room, wash their hands before and after wearing gloves, wear a gown if they anticipate direct contact with the patient or the patient's environment, limit patient transport, and use dedicated patient‐care equipment. Droplet precautions require that healthcare workers wear surgical masks when working within 3 feet of the patient and that the patient wear a surgical mask during transport. Patients under droplet precautions were placed in private rooms, when they were available, and the cohorting of such patients was permitted.
As part of our ongoing infection control program, we currently employ contact precautions to prevent transmission of several multidrug‐resistant, gram‐negative organisms and vancomycin‐resistant Enterococcus. This program includes yearly employee in‐service education. With the implementation of contact and droplet precautions during the first intervention period, the existing infection control education program for healthcare workers was expanded to include MRSA. Topics included the epidemiology and clinical features of MRSA infections, the incidence of MRSA infection at our hospital, the modes of MRSA transmission, and an explanation of contact and droplet precautions. A similar, expanded educational program was conducted for building service employees, which included discussions of environmental issues involved in controlling the spread of MRSA infection and an explanation of contact and droplet precautions. Repeated educational programs were conducted as a routine component of the hospital’s infection control program.
The second intervention period took place from May 1, 2004, through June 30, 2005, during which time droplet precautions were discontinued and only contact precautions were continued.
Statistical analysis was performed to compare changes in the rate of hospital‐acquired MRSA infection between the baseline period, the first intervention period, and the second intervention period. Because our baseline period was from March through June, and to ensure that our results were not the result of seasonal trends, we compared baseline and intervention‐period data from the same 4‐month intervals (ie, March‐June) in 2002, 2003, and 2005 (Figure 1A). We were unable to use the period from March through June 2004 as a comparator period because droplet precautions were discontinued in May 2004. Thus March through June 2005 became the next available comparator period. Statistical analyses were conducted to compare the baseline period with the first intervention period and the first intervention period with the second intervention period.
Figure 1. A, Time line for investigation of the efficacy of infection control precautions in the intensive care units (ICUs) and the periods of data analysis. B, Time line for investigation of efficacy of infection control precautions in patient care areas outside the ICUs and periods of data analysis.
Investigation in Non‐ICU Areas
A second investigation compared the effect of contact precautions with that of standard precautions in all non‐ICU areas. This investigation involved a baseline data collection period and an intervention period (Figure 1B). By use of methodologies identical to those employed in the first investigation (as described above), baseline data on MRSA infection and colonization were collected in all non‐ICU areas. The baseline period lasted from March 1, 2002, through April 30, 2004; standard precautions were used during this period for all MRSA‐positive patients. The intervention period lasted from May 1, 2004, through April 30, 2005. The single intervention in all of these patient care areas was implementation of contact precautions, as defined above, for patients who were infected or colonized with MRSA. An employee education program was included as a component of our ongoing infection control education program, as described above. A statistical analysis of changes in the rate of hospital‐acquired MRSA infection in non‐ICU areas was performed, which compared rates during the 12‐month intervention period with rates during the previous 12‐month period. We also analyzed the impact of our interventions in areas of high and low baseline transmission rates; high rates were defined as more than 1.0 infection per 1,000 patient‐days. Statistical analysis was performed using an exact 2‐tailed binomial test. All confidence levels were set at 95%. Hospital‐acquired infection rates and colonization rates were calculated as cases per 1,000 patient‐days.
Results
MRSA in the ICUs
The overall rate of hospital‐acquired MRSA infection in all ICUs during the 4‐month baseline period was 3.7 infections per 1,000 patient‐days. The combined rate of hospital‐acquired MRSA infection in the MICU and the SICU during this baseline period was 10.0 cases per 1,000 patient‐days. The combined rate for all other ICUs during the baseline period was 0.7 infections per 1,000 patient‐days. Contact and droplet precautions were implemented in all ICUs on July 1, 2002. After this intervention, the combined rate of infection in the MICU and SICU decreased from 10.0 to 4.3 cases per 1,000 patient‐days during the same 4‐month period in 2003 (95% CI, 0.17‐0.97; 
) (Figure 2). This indicated a 57% decrease in the combined rate of infection in the MICU and SICU. The infection rate for all other ICUs was 0.7 cases per 1,000 patient‐days during the baseline period and 1.0 case per 1,000 patient‐days during the first intervention period (
).
Figure 2. Rates of hospital‐acquired methicillin‐resistant Staphylococcus aureus infection in intensive care units (ICUs) during the use of standard precautions, use of contact and droplet precautions, and use of contact precautions alone. MICU, medical ICU; SICU, surgical ICU. See Methods for other ICUs analyzed.
Droplet precautions were suspended in the ICUs on May 1, 2004. Subsequently, the combined rate of hospital‐acquired MRSA infection in the MICU and SICU during the second intervention period was further reduced to 2.5 cases per 1,000 patient‐days (Figure 2). Though this decrease was a 42% improvement, compared with the first intervention period, it was not statistically significant (95% CI, 0.15‐1.91; 
).
Comparison of the combined colonization rate for the MICU and SICU during the baseline period with that during the first intervention period indicated no significant difference (1.9 vs 1.8 colonizations per 1,000 patient‐days; 
). Likewise, there was no significant difference in the colonization rates for the MICU and SICU between the first intervention period and the second intervention period (1.8 vs 1.5 colonizations per 1,000 patient‐days; 
).
An analysis of the impact of excluding patients from whom MRSA had been isolated previously was conducted. When the baseline, first intervention, and second intervention periods were compared, there was no significant difference in the number of patients excluded during each period (
).
MRSA in the Non‐ICU Areas
During the 12‐month period preceding implementation of contact precautions on May 1, 2004, the rate of hospital‐acquired MRSA infection among the non‐ICU areas was 1.1 infections per 1,000 patient‐days (Figure 3). A total of 4 inpatient care areas (general surgery, the pulmonary service, the oncology service, and the nephrology service) were identified as having high incidence rates, compared with the other non‐ICU areas evaluated (ie, the cardiology service, the pediatric service, the neurology service, and the geriatric service). The combined infection rate for the high‐incidence areas was 1.3 infections per 1,000 patient‐days, whereas the rate for the low‐incidence areas was 0.6 infections per 1,000 patient‐days. The implementation of contact precautions in non‐ICU areas led to a rate of 0.9 infections per 1,000 patient‐days in the high‐incidence areas. This was a 33% decrease from the baseline rate (95% CI, 0.47‐0.94; 
). The infection rate in low‐incidence areas did not change significantly during the intervention period (0.6 vs 0.7 infections per 1,000 patient‐days; 
).
Figure 3. Rates of hospital‐acquired methicillin‐resistant Staphylococcus aureus infection in patient‐care areas outside the intensive care units during the use of standard precautions and during the use of contact precautions alone. High incidence was defined as more than 1.0 infection per 1,000 patient‐days.
The colonization rate during the preintervention period was 1.45 cases per 1,000 patient‐days; during the intervention period, it was 0.94 cases per 1,000 patient‐days (
). This difference was not significant and indicated that fewer colonized patients were identified during the intervention period.
We conducted an analysis of the impact of excluding patients from whom MRSA had been isolated previously. During the baseline period, 33 patients were excluded from the analysis on this basis; during the intervention period, 32 were excluded (
).
Discussion
Intensive MRSA control programs in several western European countries include contact and droplet precautions, isolation of patients who are infected or colonized, and “search and destroy” strategies (ie, active surveillance for MRSA colonization in high‐risk patients and healthcare workers, followed by decolonization with antimicrobial agents).4‐6 Although these programs are effective in containing MRSA infections and in eradicating colonization, they are implemented in countries with relatively low prevalence rates of MRSA infection and well‐integrated healthcare systems. Some recommend routine implementation of similar programs in the United States.13 Many hospitals in the United States are unable or unwilling to implement similar programs, despite their demonstrated efficacy in outbreak situations.14,15 However, these studies did not examine the long‐term efficacy of these programs in reducing the rates of hospital‐acquired MRSA infection after the termination of outbreaks.
There are data demonstrating a decline in the quality of patient care as a result of the implementation of isolation precautions. Kirkland and colleagues16 found that patients who were placed under contact precautions had half as many contacts with healthcare providers, compared with patients who were not under these precautions. Additionally, certain components of the “search and destroy” approach have questionable efficacy. Nijssen et al.17 examined the efficacy of active surveillance and decolonization in reducing the incidence of hospital‐acquired infections. Their investigation found that the strains of MRSA associated with hospital‐acquired infections were genetically unrelated to the strains recovered from colonized patients.17 Although evidence supports decolonization to reduce autoinfection, there are few data supporting the long‐term implementation of surveillance, isolation, and decolonization in situations that do not involve outbreaks but in which the prevalence of MRSA is high. We undertook an investigation of specific modalities and their impact on the rate of hospital‐acquired MRSA infection in 2 distinct settings to identify an approach to MRSA containment best suited to our facility. Prior to this investigation, we relied on standard precautions to control hospital‐acquired MRSA infections. Our study was designed to implement and analyze sequential interventions on the basis of data from ongoing surveillance.
Our results consisted of several notable findings that may be validated further by prospective, simultaneously controlled studies. First, a significant decrease occurred in the rate of hospital‐acquired MRSA infection in high‐incidence ICUs when contact and droplet precautions were implemented. This decrease was not seen in low‐incidence ICUs, where the rate of infection remained unchanged. Second, although not statistically significant, the discontinuation of droplet precautions resulted in a further decrease in the rate of hospital‐acquired MRSA infection in the ICUs. Further investigation into the role of droplet precautions in reducing these infections is warranted. Third, the hospitalwide implementation of contact precautions alone significantly decreased the rate of hospital‐acquired MRSA infection in high‐incidence areas. As was the case in the ICUs, this decrease was not seen in low‐incidence areas. Unlike prior studies, our investigation was not conducted during an outbreak and included areas of high incidence and areas of low incidence of infection.
As part of our ongoing infection control program, we currently employ contact precautions for several multidrug‐resistant, gram‐negative organisms, as well as vancomycin‐resistant Enterococcus. This program includes yearly employee in‐service education. When we decided to implement contact precautions for the control of MRSA, our infection control education program was expanded to include the problem of MRSA, its modes of transmission, and strategies to interrupt transmission. This educational program is an integral part of barrier precaution implementation. Our study did not attempt to evaluate the contributory impact of this educational program. Also, we did not attempt to monitor adherence to the precautions
Several analyses were conducted to examine the potential for bias in our investigation. A comparison of colonization rates between the baseline and intervention periods did not reveal that observer bias accounted for the improved infection rates. Also, a comparison of the number of patients excluded during the baseline and intervention periods because they previously had MRSA recovered did not reveal that selection bias accounted for the improved infection rates.
We recognize that our study design was driven by the results of ongoing surveillance data and did not include control patients hospitalized during the same intervals. Our program to reduce the incidence of hospital‐acquired MRSA infections was implemented in response to unacceptable transmission rates. Data analysis was conducted to evaluate the efficacy of this program.
We did not specifically analyze the individual benefit of droplet and contact precautions in our investigations. In our ICU investigation, we compared the impact of contact and droplet precautions with the impact of contact precautions alone. This sequential methodology may not fully delineate the impact of each modality. We did note a further reduction in the rate of hospital‐acquired MRSA infection in the ICUs after droplet precautions were discontinued. This suggests a predominant role for contact transmission of MRSA in our facility and raises questions about the universal utility of droplet precautions in comprehensive MRSA infection control programs. In view of the logistic, clinical, and financial difficulty of maintaining droplet precautions, further studies are needed to better evaluate their role in reducing the incidence of hospital‐acquired MRSA infection. Despite spirited recommendations to the contrary,18 our evidence suggests that it is not obvious that extensive active surveillance is necessary to substantially limit the rate of hospital‐acquired MRSA infection. Effective contact precautions remain the most effective component of infection control efforts against nosocomial bacterial infections.
Acknowledgments
Potential conflicts of interest. All authors report no conflicts of interest relevant to this article.
Financial support. BMA Medical Foundation (Flushing, New York), Beatrice Snyder Foundation, and Agnes Varis.
References
- 1. National Nosocomial Infections Surveillance System. National Hospital‐acquired Infections Surveillance System Report, data summary from January 1992–June 2002, issued August 2002. Am J Infect Control 2002; 30:458‐75.
- 2. Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ, Karchmer AW, Carmeli Y. Comparison of mortality associated with methicillin‐resistant and methicillin‐susceptible Staphylococcus aureus bacteremia: a meta‐analysis. Clin Infect Dis 2003; 36:53‐59.
- 3. Siegel J, Rhinehart E, Jackson M, Chiarello L, the Healthcare Infection Control Practices Advisory Group. Management of multidrug‐resistant organisms in healthcare settings, 2006. Atlanta, GA: Centers for Disease Control and Prevention; 2006. Available at: http://www.cdc.gov/ncidod/dhqp/pdf/ar/mdroGuideline2006.pdf. Accessed December 10, 2006.
- 4. Pan A, Carnevale G, Catenazzi P, et al. Trends in methicillin‐resistant Staphylococcus aureus (MRSA) bloodstream infections: effects of the MRSA “search and isolate” strategy in a hospital with hyperendemic MRSA. Infect Control Hosp Epidemiol 2005; 26:127‐133.
- 5. Verhoef J, Beaujean HB, Baars A, et al. A Dutch approach to methicillin‐resistant Staphylococcus aureus. Eur J Clin Microbiol Infect Dis 1999; 18:461‐466.
- 6. Struelens M, Ronveaux O, Jans B, Mertens R, the Groupement pour le Depistage, l’Etude et la Prevention des Infections Hospitalieres. Methicillin‐resistant Staphylococcus aureus epidemiology and control in Belgian hospitals 1991 to 1995. Infect Control Hosp Epidemiol 1996; 17:503‐508.
- 7. Jernigan J, Clemence A, Stott G, et al. Control of methicillin‐resistant Staphylococcus aureus at a university hospital: 1 decade later. Infect Control Hosp Epidemiol 1995; 16:687‐696.
- 8. Muto C, Jarvis W, Farr B. Another tale of 2 guidelines. Clin Infect Dis 2006; 43:796‐797.
- 9. Strausbaugh L, Siegel J, Weinstein R. Reply to Muto et al. Clin Infect Dis 2006; 43:797‐798.
- 10. Hospital Infection Control Practices Advisory Committee (HICPAC). Management of multidrug‐resistant organisms in healthcare settings, 2006. Available at: http://www.cdc.gov/ncidod/dhqp. Accessed December 10, 2006.
- 11. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. Centers for Disease Control and Prevention definitions for nosocomial infections. In: Olmstead RN, ed. Infection Control and Applied Epidemiology: Principles and Practice. St. Louis: Mosby, 1996:A1‐A20.
- 12. Garner JS, the Hospital Infection Control Practices Advisory Committee. Guidelines for isolation precautions in hospitals. Infect Control Hosp Epidemiol 1996; 17:53‐80.
- 13. Arnold MS, Dempsey JM, Fishman M, et al. The best hospital practices for controlling methicillin‐resistant Staphylococcus aureus: on the cutting edge. Infect Control Hosp Epidemiol 2002; 23:69‐76.
- 14. Jernigan JA, Titus MG, Groschel DH, et al. Effectiveness of contact isolation during a hospital outbreak of methicillin‐resistant Staphylococcus aureus. Am J Epidemiol 1996; 143:496‐504.
- 15. Saiman L, Cronquist A, Wu F, et al. An outbreak of methicillin‐resistant Staphylococcus aureus in a neonatal intensive care unit. Infect Control Hosp Epidemiol 2003; 24:317‐321.
- 16. Kirkland KB, Weinstein JM. Adverse effects of contact isolation. Lancet 1999; 354:1177‐1178.
- 17. Nijssen S, Bonten MJ, Weinstein RA. Are active microbiological surveillance and subsequent isolation needed to prevent the spread of methicillin‐resistant Staphylococcus aureus? Clin Infect Dis 2005; 40:405‐409.
- 18. Farr B. Doing the right thing (and figuring out what that is). Infect Control Hosp Epidemiol 2006; 27:999‐1003.
-
Presented in part: 45th Interscience Conference on Antimicrobial Agents and Chemotherapy; Washington, DC; December 16‐19, 2005 (Abstract K545/416).