Reduction in Hospitalwide Incidence of Infection or Colonization with Methicillin‐Resistant Staphylococcus aureus With Use of Antimicrobial Hand‐Hygiene Gel and Statistical Process Control Charts
Objective. To evaluate the impact of serial interventions on the incidence of methicillin‐resistant Staphylococcus aureus (MRSA).
Design. Longitudinal observational study before and after interventions.
Setting. The Alfred Hospital is a 350‐bed tertiary referral hospital with a 35‐bed intensive care unit (ICU).
Interventions. A series of interventions including the introduction of an antimicrobial hand‐hygiene gel to the intensive care unit and a hospitalwide MRSA surveillance feedback program that used statistical process control charts but not active surveillance cultures.
Methods. Serial interventions were introduced between January 2003 and May 2006. The incidence and rates of new patients colonized or infected with MRSA and episodes of MRSA bacteremia in the intensive care unit and hospitalwide were compared between the preintervention and intervention periods. Segmented regression analysis was used to calculate the percentage reduction in new patients with MRSA and in episodes of MRSA bacteremia hospitalwide in the intervention period.
Results. The rate of new patients with MRSA in the ICU was 6.7 cases per 100 patient admissions in the intervention period, compared with 9.3 cases per 100 patient admissions in the preintervention period (
). The hospitalwide rate of new patients with MRSA was 1.7 cases per 100 patient admissions in the intervention period, compared with 3.0 cases per 100 patient admissions in the preintervention period (
). By use of segmented regression analysis, the maximum and conservative estimates for percentage reduction in the rate of new patients with MRSA were 79.5% and 42.0%, respectively, and the maximum and conservative estimates for percentage reduction in the rate of episodes of MRSA bacteremia were 87.4% and 39.0%, respectively.
Conclusion. A sustained reduction in the number of new patients with MRSA colonization or infection has been demonstrated using minimal resources and a limited number of interventions.
Received October 9, 2006; accepted December 15, 2006; electronically published May 31, 2007.
Methicillin‐resistant Staphylococcus aureus (MRSA) remains a significant problem for Australian hospitals. In 2004, the Australian Group for Antimicrobial Resistance found that 14.9% of 2,652 S. aureus isolates from 27 hospitals and private laboratories were methicillin resistant.1 Between January 1999 and December 2002, 51% of cases of S. aureus bacteremia were hospital acquired (ie, had onset more than 48 hours after admission), and 40% were due to methicillin‐resistant strains.2 At our institution in 2000‐2001, 6.8% of 1,185 patients in the intensive care unit (ICU) were colonized or infected with MRSA at admission, and 11.4% of 554 patients acquired MRSA during their ICU stay.3
There is considerable debate in the literature regarding the optimal infection control measures to control the spread of MRSA.4‐10 Strategies range from standard precautions to use of active surveillance cultures with contact precautions and multimodal, hospitalwide hand‐hygiene intervention programs that include use of alcohol‐based hand rubs. Multimodal intervention programs incorporated additional interventions, such as observational compliance audits with performance feedback, educational and promotional activities, mupirocin decolonization, antiseptic body washes, and enhanced equipment cleaning.11,12 Recently, the Centers for Disease Control and Prevention (CDC) Healthcare Infection Control Practice Advisory Committee (HICPAC) issued guidelines for the management of multidrug‐resistant organisms in healthcare settings, which recommend a 2‐tiered approach whereby strategies at a local level can be escalated or de‐escalated subject to the effectiveness of interventions, the incidence, the prevalence, or the emergence of epidemiologically important multidrug‐resistant organisms.13
Use of active surveillance cultures combined with contact precautions and multimodal, hospitalwide hand‐hygiene programs has been shown to be economically beneficial; however, these interventions require substantial resources to introduce and maintain.11,12,14 Another novel and less costly approach shown to control the spread of MRSA is a surveillance feedback program using annotated statistical process control (SPC) charts.15 SPC combines rigorous time series analysis methods with timely presentation of data in a graphical format that is easily understood by healthcare workers who may have limited understanding of statistics.16,17 Repeated measurements from any process will often display natural variation.18 SPC charts help to distinguish between “common cause variation,” that is, data that are stable and predictable, and “special cause variation,” that is, data that are caused by irregular or unnatural causes. Processes that exhibit common cause variation are said to be “in statistical control,” and processes that exhibit special cause variation are said to be “out of control” due to a special cause, such as specific interventions or other factors beyond our control.16
As in other countries, inadequate compliance with hand‐hygiene practices compounds the MRSA problem in Australia.19 The average level of adherence to hand‐hygiene guidelines is estimated to be less than 50%, although this varies between clinical settings, healthcare worker disciplines, and intensity of workload.19 Noncompliance has been highest in ICU areas.20,21
To minimize the impact on infection control resources and costs associated with strategies to control MRSA, infection control at the Alfred Hospital implemented a program with a limited number of interventions. These included the introduction of a gel‐based antimicrobial hand‐hygiene product into the ICU for fast‐acting, waterless hand antisepsis and an MRSA surveillance feedback program using SPC charts in the ICU and individual wards hospitalwide. We aimed to determine whether these interventions led to lower numbers of new patients colonized or infected with MRSA (hereafter, “new patients with MRSA”) and episodes of MRSA bacteremia, compared with those in the preintervention period.
Methods
Setting and Design
The study was conducted at the Alfred Hospital, which is a 350‐bed tertiary referral hospital with a 35‐bed ICU. The study was an interrupted time series comparing the preintervention period with the intervention period. The study period for the ICU and hospitalwide was from January 2001 to May 2006. The preintervention period was from January 2001 to December 2002, and the intervention period was from January 2003 to May 2006.
Interventions
Table 1 summarizes the serial interventions and the intervention time periods.
Introduction of antimicrobial hand‐hygiene gel. In January 2003, because of additional funding and after approval from the hospital ethics committee, an antimicrobial hand‐hygiene gel not registered in Australia (Sterigel+; Les Enterprises Solumed) containing chlorhexidine gluconate (0.5% wt/vol) and ethyl alcohol (70% vol/vol) was introduced into the ICU. Because of Australian Therapeutic Goods Administration (TGA) restrictions, this product was available only for a limited period (Table 1). Standard hand‐hygiene products remained available in the ICU and included an antiseptic hand‐washing agent, Bioprep (Ecolab), containing 4% chlorhexidine gluconate, and a hand rub, chlorhexidine 1% hand lotion (Orion Laboratories), containing 10 g/L chlorhexidine gluconate and 70% v/v ethanol. A supply of four 500‐mL bottles of Sterigel+ was maintained in each ICU cubicle or bed space. The infection control team distributed this hand gel and maintained a database on usage between January 2003 and September 2003.
Sign for antibiotic‐resistant organisms. During the same period, any ICU patient in which MRSA was identified had a sign stating “antibiotic‐resistant organism” placed in the cubicle area (Table 1). In September 2003, Sterigel+ was withdrawn from the ICU in accordance with TGA restrictions, and the posting of signs for antibiotic‐resistant organisms also ceased.
In late November 2003, an alternative Australian‐registered antimicrobial hand‐hygiene gel, Microshield (Johnson and Johnson), containing ethanol (61.5% v/v), replaced the standard hand rub (chlorhexidine 1% hand lotion) in the ICU (Table 1). The number of bottles supplied to each ICU cubicle remained the same as for Sterigel+, and Microshield gel was distributed via the Pharmacy Department delivery system. There was no change in the standard hand‐hygiene products available hospitalwide. Bioprep and the chlorhexidine 1% hand lotion remained available in all wards.
Feedback program using SPC charts. In June 2004, additional funding allowed the establishment of an MRSA surveillance feedback program using SPC charts throughout the hospital (including the ICU) (Table 1). By use of 24 months of historical data on MRSA colonization and infection from the microbiology laboratory database, control charts were developed for the ICU, 15 inpatient wards, and the entire hospital.
To assist ward staff with interpretation, the mean was labeled as the “average,” and 2‐SD and 3‐SD control limits above the mean were labeled as the “warning limit” and “action limit,” respectively. If the number of new patients with MRSA per month reached the warning limit, ward staff were advised and were asked to investigate. If the number per month reached the action limit, this was treated as a possible “special cause variation,” and the infection control team conducted an investigation. Feedback was given monthly. Intensivists and ICU nurse managers received the SPC chart for MRSA in the ICU; ward nurse managers received SPC chart for MRSA in their own ward, the hospitalwide SPC chart for MRSA, and a table summarizing the new patients with MRSA in their ward during the previous month. Of the 15 wards that historically had the highest number of new patients with MRSA per year, 5 were targeted for additional feedback. Each month, staff from these wards met with an infection control practitioner to discuss the SPC charts and any new patients with MRSA. In addition, the heads of the units with patients in these wards received a monthly written report that included the SPC charts for MRSA and a table summarizing clinical information on each new patient with MRSA. The SPC charts for MRSA were also displayed on notice boards in the ICU and the target wards.
In‐service education in the intervention period was limited to a 2‐week period before the introduction of the first, nonregistered antimicrobial hand‐hygiene gel and a 2‐week period before the commencement of the SPC MRSA‐surveillance feedback program. Education sessions focused on instructions for the appropriate use of this gel and on briefing the nurse managers and ward staff about the interpretation of SPC charts for MRSA. The Public Relations Department periodically reported on outcomes in the hospital newsletter.
Process Measures
Usage of the hand‐hygiene product. Usage of the hand‐hygiene product was used as a proxy for hand‐hygiene practices and was calculated as the volume in liters of hand‐hygiene product distributed to the ICU per 1,000 patient‐days. Distribution data for standard hand‐hygiene products were provided by the Pharmacy Department and the hospital contract cleaning service.
Usage of the hand‐hygiene product was measured in the preintervention period between June 2002 and December 2002 and in the intervention period between January 2003 and September 2003 and between August 2005 and October 2005. Formal observations of compliance with guidelines for hand hygiene were not made.
Signs for antibiotic‐resistant organisms. A sign stating “antibiotic‐resistant organism” was displayed in the cubicle area of any patient in the ICU in whom MRSA had been identified. Opportunities to display the sign in the ICU were measured in the intervention period between January 2003 and September 2003.
Outcome Measures
The outcome measures for the study were the numbers of new patients with MRSA and episodes of MRSA bacteremia. We evaluated changes in rates and incidence of new patients with MRSA and episodes of MRSA bacteremia, using patient‐days, patient admissions, and device‐days as the units for denominators. Active surveillance cultures were not performed during the study period.
New patient with MRSA. A new patient with MRSA was defined as any inpatient in whom MRSA was detected for whom no known prior history of MRSA could be established. This definition included patients who were infected and/or colonized. Patients were counted only once, and duplicate isolates were not included.
MRSA bacteremia. An episode of MRSA bacteremia was defined on the basis of an MRSA‐positive blood culture from any inpatient. If, for the same patient, a 14‐day period had passed between one MRSA‐positive blood culture and the second positive culture result, this was considered to indicate a new episode, which was included in the data set.
Rates of central line–associated bloodstream infection (CLABSI) in the ICU. CLABSI and MRSA CLABSI in the ICU were defined using the standardized CDC definitions, and rates were calculated per 1,000 central line device–days.22
Routine Infection Control Precautions for MRSA
Where possible, patients with MRSA colonization or infection should be housed in a single room or with patients who do not have relevant risk factors (such as immunosuppression, invasive devices, or surgical wounds), and staff must follow standard precautions. For multiple reasons, single rooms are rarely available; hence, patients with MRSA colonization or infection are cared for in rooms where a bed can be found.
Intensified Infection Control Strategies and Precautions in the ICU during Outbreaks
During the study period and in response to an increase in rates of CLABSI in the ICU after the introduction of a needleless intravenous system with a mechanical valve (November 2001) and an outbreak of Acinetobacter infection (October 2002‐May 2006), escalated infection control strategies and precautions were implemented. These included the use of antiseptic‐ and antibiotic‐coated central venous catheters, starting in February 2003. In response to the outbreak of Acinetobacter infection, the escalated precautions implemented included glove use upon entry into the cubicle area of all ICU patients from October 2002 to November 2002, use of gowns and gloves upon entry into all ICU patient cubicles from June 2004 to February 2005, and use of gowns and gloves upon entry into all ICU patient cubicles by all staff moving from cubicle to cubicle and the use of standard precautions by the primary nurse, starting in February 2005.
Statistical Analysis
Process and outcome measures were compared between the preintervention and intervention periods for the ICU and hospitalwide by use of the Wilcoxon rank‐sum test and independent group t test, as appropriate, with STATA, version 8 (Stata). A 2‐sided P value of .05 was considered to be statistically significant.
Shewhart “c” control charts were used because the data consisted of a count of new MRSA isolates. Control chart limits were set at 3 SDs of the mean.15 A shift was defined as 8 or more consecutive data points on the same side of the mean.15,16
Segmented regression analysis of interrupted time series was used to compare the preintervention and intervention periods for new patients with MRSA and episodes of MRSA bacteremia hospitalwide and was performed using SAS, version 8.2 (SAS Institute). Changes over time and maximum and conservative estimates were determined using the autoregression procedure in SAS, which adjusts the regression process for autocorrelation between time points.23,24 Durbin‐Watson statistics and partial autocorrelation functions were used to check for possible autocorrelation.
Results
The results of process and outcome measures are summarized in Table 2.
Usage of Hand‐Hygiene Product
Distribution data for standard hand‐hygiene products used in the ICU were available for a 7‐month period before the commencement of the interventions in the ICU (June 2002‐December 2002). The overall rate of usage of the standard agents in the preintervention period was 78.1 L per 1,000 patient‐days. During the first 9 months of the intervention period (January 2003‐September 2003), the total overall rate of usage of all products, including Sterigel+, was 102.7 L per 1,000 patient‐days, an increase of 32%. The usage of standard agents in the ICU remained unchanged from the preintervention period.
Ongoing product‐usage data were not collected after September 2003. However, a review of the ICU distribution data for 3 months (August 2005‐October 2005) demonstrated that the standard agent usage rate was 120.2 L per 1,000 patient‐days, and Microshield usage rate was 82 L per 1,000 patient‐days, suggesting sustained hand‐hygiene practices.
Sign for Antibiotic‐Resistant Organisms
During the first 9 months of the intervention period (January 2003‐September 2003), 153 new patients were found to harbor MRSA while in the ICU. The sign for antibiotic‐resistant organisms was unable to be displayed for 85 (56%) of the 153 patients, because 64 had been discharged to a ward and 8 were deceased at the time the results became available. The remaining 13 patients did not have signs displayed for multiple reasons, including laboratory notification of MRSA status after September 2003.
New Patients with MRSA
The rate of new patients with MRSA in the ICU in the intervention period was 6.7 cases per 100 patient admissions (95% CI, 5.2‐8.5), compared with 9.3 cases per 100 patient admissions (95% CI, 7.5‐11.2) in the preintervention period (
). The hospitalwide rate of new patients with MRSA was 1.7 cases per 100 patient admissions (95% CI, 1.6‐1.8) in the intervention period, compared with 3.0 cases per 100 patient admissions (95% CI, 2.8‐3.2) in the preintervention period (
).
CLABSI Rates in the ICU
Data on rates of CLABSI in the ICU were available from June 2001 to December 2002 in the preintervention period. The overall rate of CLABSI in the ICU in the preintervention period was 12.1 cases per 1,000 device‐days (95% CI, 10.3‐14.2), compared with 11.1 cases per 1,000 device‐days (95% CI, 9.7‐12.6) in the intervention period (
). The MRSA CLABSI rates in the ICU in the preintervention period was 2.0 cases per 1,000 device‐days (95% CI, 1.3‐3.0), compared with 1.1 cases per 1,000 device‐days (95% CI, 0.7‐1.7) in the intervention period (
).
MRSA Bacteremia
The hospitalwide rate of episodes of MRSA bacteremia was 0.27 episodes per 100 patient admissions (95% CI, 0.24‐0.32) in the intervention period, compared with 0.45 episodes per 100 patient admissions (95% CI, 0.38‐0.52) in the preintervention period (
).
SPC Charts
Interpretation of the ICU and hospitalwide SPC charts for new patients with MRSA demonstrated 8 or more consecutive data points on the same side of the mean, indicating a shift in the process and a reduction in the number of new patients with MRSA in the intervention period (Figures 1 and 2). It appeared that the hospitalwide incidence of new patients with MRSA started to decline during the ICU‐specific interventions (Figure 2). During October and November 2003, the number of new patients with MRSA in the ICU increased when all interventions (including the use of Sterigel+) ceased. When Microshield was introduced in late November 2003, the number of new MRSA isolates declined.
Figure 1. Statistical process control (SPC) chart showing the number of new patients colonized or infected with methicillin‐resistant Staphylococcus aureus (MRSA) in the intensive care unit (ICU) per month during the preintervention and intervention periods and the interventions that were implemented.
Figure 2. Statistical process control (SPC) chart showing the number of new patients colonized or infected with methicillin‐resistant Staphylococcus aureus (MRSA) hospitalwide per month during the preintervention and intervention periods and the interventions that were implemented.
Segmented Regression Analysis
By segmented regression analysis, there was a significant decrease in the level and trend (slope) of both new patients with MRSA and episodes of MRSA bacteremia in the intervention period (Figures 3 and 4). The rate of new patients with MRSA decreased by 0.68 cases per 100 patient admissions per month (95% CI, 0.14‐1.22; 
), and the rate of MRSA bacteremia decreased by 0.16 cases per 100 patient admissions per month (95% CI, 0.01‐0.31; 
) (Table 3). The trend in the rate of new patients with MRSA decreased by 0.07 cases per 100 patient admissions per month (95% CI, 0.03‐0.10; 
), and the trend in the rate of MRSA bacteremia decreased by 0.02 cases per 100 patient admissions per month (95% CI, 0.03‐0.01; 
) (Table 3).
Figure 3. Segmented regression analysis showing the number of new patients with methicillin‐resistant Staphylococcus aureus (MRSA) infection or colonization per 100 patient admissions hospitalwide in the preintervention period (triangles) and the intervention period (circles). Month 1 = January 2001.
Figure 4. Segmented regression analysis showing the number of episodes of methicillin‐resistant Staphylococcus aureus (MRSA) bacteremia per 100 admissions hospitalwide in the preintervention period (triangles) and the intervention period (circles).
We estimated that the maximum reduction effect of the interventions on the rate of new patients with MRSA was a decrease of 3.34 cases per 100 admissions per month and estimated that the conservative reduction effect was a decrease of 1.76 cases per 100 admissions per month. The estimated maximum and conservative reduction effects on the rate of MRSA bacteremia was a decrease of 0.94 cases per 100 admissions per month and 0.42 cases per 100 admissions per month, respectively (Table 3). The maximum and conservative estimates for percentage reduction in the rate of new patients with MRSA were 79.5% and 42.0%, respectively, and the maximum and conservative estimates for percentage reduction in the rate of episodes of MRSA bacteremia were 87.4% and 39.0%, respectively. The Durbin‐Watson statistics for new patients with MRSA and episodes of MRSA bacteremia were 1.955 and 2.018, respectively, indicating no serial autocorrelation.
Discussion
The serial interventions in this study were followed by a significant reduction in the rates of new patients with MRSA and episodes of MRSA bacteremia hospitalwide. This was demonstrated in a setting in which active surveillance cultures were not used.
Many studies of control strategies for MRSA have been conducted during outbreaks and have included the introduction of multiple interventions concurrently; hence, determining the relative contribution of each intervention has not been possible.9 Our study was conducted in over an extended period in a setting of high endemicity but no outbreak, with a more limited number of interventions than in other published reports, which may help to identify the specific interventions responsible for the reduction in the rate of MRSA colonization and infection. The timely use of a sign indicating isolation of antibiotic‐resistant organisms proved not to be possible for many patients who acquired MRSA in the ICU, and we concluded that this intervention was likely to have had little impact on the reduction in the rate of MRSA colonization and infection.
There may have been other unidentified factors responsible for the reduction, and we can report only an association between the interventions and the decrease in the rate of MRSA colonization and infection; however, analysis with both SPC charts and segmented regression support that these interventions are responsible for the reduction. The occurrence of 8 consecutive data points on the same side of the mean for the SPC charts met the “rules” used to interpret SPC charts and suggest that the shift was not random variation and was a real, sustained reduction in the number of new patients with MRSA. Use of segmented regression analysis may overcome some of the problems in analyzing preintervention and postintervention data, including lack of independence of outcomes and natural variation in rates of MRSA colonization and infection.
Consistent with reported findings, the use of SPC charts at our institution created awareness about MRSA and devolved responsibility to the local ward or unit level which, in turn, decreased the workload of the infection control team.15 Targeting wards with historically high rates of MRSA colonization and infection gave staff in these areas an opportunity to discuss the cases in detail, to devise their own improvement strategies to decrease the risk of MRSA acquisition, and to improve infection control practices in general, which suggests that SPCs with feedback are a useful strategy to aid and stimulate a climate of culture change.
One of the limitations of this study was our inability to control for nonstudy interventions that were implemented in response to an increase in rates of CLABSI and to an outbreak of Acinetobacter infection in the ICU. These strategies may have contributed to the reduction in the number of new patients with MRSA in the ICU. Although there was a reduction in the rates of MRSA CLABSI during the intervention period, the overall rate of BSI remained high and was not statistically significantly different in the intervention period. Strategies to control the outbreak of Acinetobacter infection were intermittent, and compliance was variable. During the periods when these strategies were not in place, the rate of new patients with MRSA continued to decrease in the ICU.
To implement and oversee the running of the program, a dedicated infection control practitioner was employed for 20 hours per week (at AU$38,900 per year). These costs were substantially less than the costs associated with implementing and sustaining programs that include aggressive control measures, use of active surveillance cultures, and multimodal hand‐hygiene intervention programs.8,11,12
There are very valid arguments that the costs associated with aggressive intervention strategies to reduce and control MRSA infection can be offset by the cost savings associated with infections that will be prevented.25,26,27 Despite this demonstrated cost‐effectiveness, the reality in many countries, including Australia, is that such up‐front funding may not be readily available, since Australian public hospitals have exceeded their budgets in recent years,28 which has resulted in significant financial constraints and cost cutting.
We have shown that the number of new patients with MRSA and hospitalwide rates of MRSA bacteremia can be reduced in a setting in which MRSA is highly endemic without the use of active surveillance cultures and isolation measures. We have shown a sustained reduction in the number of new patients with MRSA, using minimal resources and a limited number of interventions. Our approach is consistent with the newly released CDC HICPAC guidelines, which include monitoring trends, implementing and measuring the impact of interventions, and escalating or intensifying strategies only if there has been no response to the interventions.13 Ongoing research to evaluate the long‐term use of less costly, less time‐consuming, targeted approaches is required to determine the most‐effective methods for prevention and control of MRSA infection.
Acknowledgments
We are grateful to Evonne Curran for her assistance with the use of SPC charts in the healthcare setting. We are grateful for the support this program has received from intensivists, ICU nurse managers, ICU nursing staff, ward nurse managers, and ward nursing staff.
Financial support. This research was supported in part by Les Entreprises Solumed, Medical Specialities Australia, and Ansell Limited.
Potential conflicts of interest. All authors report no conflicts of interest relevant to this article.
References
- 1. Nimmo GR, Coombs GW, Pearson JC, et al. Methicillin‐resistant Staphylococcus aureus in the Australian community: an evolving epidemic. Med J Aust 2006; 184:384‐388.
- 2. Collignon P, Nimmo GR, Gottlieb T, Gosbell IB. Staphylococcus aureus bacteraemia, Australia. Emerg Infect Dis 2005; 11:554‐561.
- 3. Marshall C, Harrington G, Wolfe R, Fairley CK, Wesselingh S, Spelman D. Acquisition of methicillin‐resistant Staphylococcus aureus in a large intensive care unit. Infect Control Hosp Epidemiol 2003; 24:322‐326.
- 4. Farr BM, Bellingham G. Pro/con clinical debate: isolation precautions for all intensive care unit patients with methicillin‐resistant Staphylococcus aureus colonisation are essential. Crit Care 2004; 8:153‐156.
- 5. Muto CA. Methicillin‐resistant Staphylococcus aureus control: we didn’t start the fire, but it’s time to put it out. Infect Control Hosp Epidemiol 2006; 27:111‐115.
- 6. Sunenshine R, Liedtke LA, Fridkin SK, Strausbaugh LJ. Management of inpatients colonized or infected with antimicrobial‐resistant bacteria in hospitals in the United States. Infect Control Hosp Epidemiol 2005; 26:138‐143.
- 7. Farr BM. Prevention and control of methicillin‐resistant Staphylococcus aureus infections. Curr Opin Infect Dis 2004; 17:317‐322.
- 8. Boyce JM, Havill NL, Kohan C, Dumingan DG, Ligri CE. Do infection control measures work for methicillin‐resistant Staphylococcus aureus? Infect Control Hosp Epidemiol 2006; 27:116‐121.
- 9. Cooper BS, Kibbler CC, Cookson BD, et al. Isolation measures in the hospital management of methicillin‐resistant Staphylococcus aureus (MRSA): systematic review of the literature. BMJ 2004; 329:533‐539.
- 10. Farr B. What to think if results of the National Institutes of Health randomised trial of methicillin‐resistant Staphylococcus aureus and vancomycin‐resistant Enterococcus control measures are negative (and other advice to young epidemiologists): a review and an au revoir. Infect Control Hosp Epidemiol 2006; 26:1098‐1106.
- 11. Pittet D, Hoougonnet S, Harbarth S, et al. Effectiveness of a hospital‐wide program to improve compliance with hand hygiene. Lancet 2000; 356:1307‐1312.
- 12. Johnson PD, Martin R, Burrell LJ, et al. Efficacy of an alcohol/chlorhexidine hand hygiene program in a hospital with high rates of nosocomial methicillin‐resistant Staphylococcus aureus (MRSA) infection. Med J Aust 2005; 183:509‐514.
- 13. Siegel JD, Rheinhart E, Jackson M, Chiarello L, the Healthcare Infection Control Practice Advisory Committee. Management of Multi‐Resistant Organisms in Healthcare Settings, 2006. Available at: http://www.cdc.gov/ncidod/dhqp/pdf/ar/mdroGuideline2006.pdf. Accessed November 12, 2006.
- 14. Humphrey H. Control of methicillin‐resistant Staphylococcus aureus in hospitals: an impossible dream? J Med Microbiol 2002; 51:283‐285.
- 15. Curran ET, Benneyan JC, Hood J. Controlling methicillin‐resistant Staphylococcus aureus: a feedback approach using annotated statistical process control charts. Infect Control Hosp Epidemiol 2002; 23:13‐18.
- 16. Benneyan JC, Lloyd RC, Plsek PE. Statistical process control as a tool for research and health care improvement. Qual Saf Health Care 2003; 12:458‐464.
- 17. Lloyd RC. Quality Health Care: A Guide to Developing and Using Indictors. 1st ed. Boston, MA: Jones and Bartlett; 2004:216.
- 18. Wadsworth HM. Statistical process control. In: Juran AM, Godfrey AB. Juran’s Quality Handbook. 5th ed. McGraw‐Hill: Singapore; 2000:45.1‐46.1.
- 19. Tibballs J. Teaching hospital medical staff to handwash. Med J Aust 1996; 164:14‐17.
- 20. Pittet D. Hand hygiene: improved standards and practice for hospital care. Curr Opin Infect Dis 2003; 16:327‐335.
- 21. Pittet D, Mourouga P, Perneger TV, Members of the Infection Control Program. Compliance with handwashing in a teaching hospital. Ann Intern Med 1999; 130:126‐130.
- 22. Horan TC, Gaynes RP. Surveillance of nosocomial infections. In: Mayhall CG, ed. Hospital Epidemiology and Infection Control. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2004:1659‐1702.
- 23. Wagner AK, Soumerai SB, Zhang F, Ross‐Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 2002; 27:299‐309.
- 24. Ansari F, Gray K, Nathwani D, et al. Outcome of an intervention to improve hospital antibiotic prescribing: interrupted time series with segmented regression analysis. J Antimicrob Chemother 2003; 52:842‐848.
- 25. Chaix C, Durand‐Zaleski I, Alberti C, Brun‐Buisson C. Control of endemic methicillin‐resistant Staphylococcus aureus: a cost‐benefit analysis in an intensive care unit. JAMA 1999; 282:1745‐1751.
- 26. Kim T, Oh PI, Simor AE. The economic impact of methicillin‐resistant Staphylococcus aureus in Canadian hospitals. Infect Control Hosp Epidemiol 2001; 22:99‐104.
- 27. West TE, Guerry C, Hiott M, et al. Effect of targeted surveillance for control of methicillin‐resistant Staphylococcus aureus in a community hospital system. Infect Control Hosp Epidemiol 2006; 27:233‐238.
- 28. Noble T. The year ahead: health. The Age. January 3, 2004. Available at: http://www.theage.com.au/articles/2004/01/02/1072908911917.html. Accessed September 25, 2006.
-
Presented in part: Hong Kong Infection Control Nurses’ Association, 2nd International Infection Control Conference; Hong Kong; June 16‐18, 2006 (Abstract 18); and the Australian Infection Control Association, 4th Biennial National Conference; Sydney, Australia; September 20‐22, 2006 (p. 39).