Divided Lines

In January, the Society for Healthcare Epidemiology of America (SHEA) and the Association of Professionals in Infection Control and Epidemiology (APIC) released a position statement regarding the use of active surveillance cultures to screen for methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), and other multi-drug resistant organisms (MDROs). This statement was released one year after two states attempted to pass legislation mandating hospitals to screen for MRSA and other MDROs. The proposed legislation is as follows:

In Illinois 

Two bills were proposed in Illinois. According to state records, SB2771 amends the University of Illinois Hospital Act and the Hospital Licensing Act. It states that any hospital subject to either of the acts must screen all patients for MRSA in accordance with guidelines published by the Centers for Disease Control and Prevention (CDC). The bill adds that if any screened patient tests positive for the organism, the hospital must inform the patient and offer treatment.

In addition, the bill would require the states hospitals to report all MRSA cases to the State Department of Public Health. SB2771 was proposed by Sen. Christine Radogno (R) in Jan. 2006, but never went further than its first read.

SB3087 generally requires the same screening mandates, but additionally requires that all patients who test positive for MRSA be segregated from patients who test negative, and the hospital must provide treatment to the MRSA-positive patient. This bill was proposed by Sen. John J. Cullerton (D) and Sen. Susan Garrett (D). It too was filed in Jan. 2006, and did get slightly further than the first bill (SB2771), however, in February it was postponed and has since adjourned sine die.

In Maryland 

HB966 was proposed in Maryland requiring hospitals and nursing facilities to establish an infection control (IC) program that works in conjunction with the Department of Health and Mental Hygiene (DHMH) and other specified groups and stakeholders to report specified healthcare acquired infections (HAIs). This bill was submitted by Delegate James W. Hubbard in Feb. 2006, and was killed in March after its first hearing.

The initiation of such actions by these two state legislations was enough to cause both APIC and SHEA to take an outspoken stand on the topic. The two note, It must be acknowledged that the discussion generated by the proposed legislation represents a critical opportunity to further raise and sustain the profile of antimicrobial resistance as a public health crisis and to better inform the public about this threat.¹ 

The SHEA and APIC position statement, Legislative mandates for use of active surveillance cultures to screen for methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci: position statement from the Joint SHEA and APIC Task Force, includes the following five highlighted points on the topic:

1. They (APIC and SHEA) do not support legislation to mandate the use of active surveillance cultures for screening 
2. They do support the continued development, validation, and application of efficacious and cost effective strategies for prevention 
3. They welcome efforts by healthcare consumers, and private, local, state, and federal policy makers, to focus attention on and formulate solutions for the growing problems encircling this topic 
4. They support ongoing additional research 
5. They support stronger collaboration between state and local public health authorities and institutional infection prevention and control experts 

In their overview of the literature on the topic, SHEA and APIC state, There is considerable evidence to support the use of active surveillance cultures for high risk patients and during outbreaks of infection and colonization with antimicrobial resistant pathogens, as has been previously recommended by SHEA and HICPAC. However, at present, there is insufficient evidence to justify the mandatory application of this strategy to all hospitalized patients.

The Preliminaries

In 1976, a nationwide survey found that 87 percent of hospitals practiced some form of infection surveillance half of which reported very active programs and routine culturing was found more prevalent in hospitals using passive surveillance methods.² Although the surveyors specifically note a boost in IC practices and programs from 1970 to 1976, this survey indicated that the rate of routine environmental cultures were already beginning to decline by about 25 percent.

Many challenges and arguments can be presented when broaching the topic of surveillance measures. Is it truly beneficial? How can it be cost-effective? How can the flow of current practices be shifted to allow for such studious and all-inclusive activities?

The first place to begin with such a program is to examine the theoretical vs. actual implementation of the program. Discrepancies between the two, when referring to the implementation of a surveillance program, can vary widely, according to a study published in the April issue of the American Journal of Infection Control. The researchers note that surveillance programs can be hindered by many factors, some of which may include: methodologies not uniformly applied; varying management, analysis, and reporting of data; infrequent or irregular infection control practitioner (ICP) hours; timely and accurate laboratory reporting; report distribution to key persons; specific actions taken in response to the process and/or the outcome of the reports.³ Other important steps to take when planning any such uniform intervention may include preparing the laboratory for the influx of work and reducing the turnaround time for the screening tests; monitoring and optimizing compliance of new contact precautions measures; educating all staff on the known adverse effects of noncompliance of the program steps (i.e. contact precautions); and properly measuring the important outcomes to evaluate the effectiveness of the program.4

The quality of the surveillance data is also crucial. This very aspect was assessed in a recent Australian study when researchers assessed the accuracy of the data from a MRSA surveillance program that included investigating accompanying bloodstream infections (BSIs). Their results were recorded as follows:5

For reported MRSA infection, estimated values were: sensitivity - 40 percent; specificity - 99.9 percent; and positive predictive value - 33.3 percent.

For reported BSIs, the estimated values were: sensitivity - 42.9 percent; specificity - 99.8 percent; and positive predictive value - 37.5 percent.

Gaining Importance and Changing Face

MRSA is no longer confined to institutional walls. In fact, even the strains that initially were tied to community-acquired are now causing disease in healthcare settings.6

Within the walls of the institution, these insidious organisms lurk. And they arent only transferable by human contact. In fact, being assigned a room where the prior occupant was infected by MRSA or VRE significantly raises a patients odds of contracting a like infection.

This hypothesis was tested by the departments of medicine and infection control at Brigham and Womens Hospital and Harvard Medical School when researchers studied the prevalence of such environmental contamination with MRSA and VRE in patient rooms.7 They assessed the relative odds of acquiring an infection with either organism by simply populating a room where the prior patient had tested positive for either type of organism. The researchers found a 1 percent increase in MRSA infection among patients whose prior room occupant was MRSA-positive compared to those whose prior room occupant was MRSA negative. Patients with prior VRE-positive occupants were at a nearly double risk of acquiring the bug.

The scientists note that the excess risks accounted for 5.1 percent of all incident MRSA cases and 6.8 percent of all incident VRE cases. Interestingly, acquisition was significantly associated with longer post ICU (intensive care unit) length of stay in this study.

Targeted Surveillance

In Germany, active surveillance and admission screening are recommended to control the spread of MRSA.8 German healthcare professionals have found that screening all patients admitted to ICU is cost effective when the incidence of MRSA is high (they note high as greater than two cases per 100 patients for community-acquired MRSA [CA-MRSA] and 0.3 infections per 100 patients for nosocomial-associated MRSA infections).

In addition, active surveillance culturing prevents further spread of VRE, according to researchers in the department of epidemiology and preventive medicine at the University of Maryland School of Medicine, Baltimore. In fact, the researchers found a significant increase in hospital-wide incidence rates when active surveillance was discontinued, and they note a significant decrease in incidence rates when it was again reinstated.9

Another way surveillance cuts down on infection transmission is through its association with contact precaution measures. Washington University School of Medicine and Barnes Jewish Hospital, both located in St. Louis, Mo., investigated the impact of an active surveillance program implemented for MRSA on contact precaution utilization.10 Six percent, or 214, of the 3,461 total contact precaution days in the ICU were attributable to the MRSA active surveillance program. The implementation of rapid, same-day results for MRSA active surveillance increased contact precaution days by 15 percent compared with no surveillance, they found. Active surveillance increased total contact precaution days in the ICU studied detected 58 percent of MRSA cases that would have otherwise been missed.

While surveillance is noted to significantly increase the detection of MRSA, this benefit is not necessarily uniform across ICUs and hospitals even with high compliance and the use of correct denominators, according to researchers at Brigham and Womens Hospital, Boston.11 They found such variances in surveillance programs can provide anywhere from a 7 percent to 157 percent increase in the detection of MRSA infection. Moreover, the reporting of incidence using the total number of patients or total patient days, can underestimate the actual incidence by a full one-third. The Brigham scientists also note that admission surveillance provided a 30 percent to 135 percent increase in rates of MRSA detection in the 140 ICUs they studied.

The cost associated with surveillance programs are generally the No. 1 initial consideration at least by the institutions administration. A cost analysis and the effect of targeted MRSA surveillance on the rate of nosocomial-related MRSA infection in a community hospital system was studied in 2006.12 This group of scientists compared the rate of MRSA infection before and after the initiation of their program. Cost effectiveness in this model was calculated as the difference between the cost savings associated with preventing nosocomial-related MRSA infections and surgical site infections (SSIs) after the cost of MRSA cultures and contact isolation for patients found colonized with MRSA. Two separate hospitals participated in the study.

The surveillance model was found to be cost effective, according to the researchers, because it prevented 13 nosocomial-related MRSA infections and nine SSIs. The estimated savings tied to those cases was $1,545,762 in 2002 U.S. dollars.

As a result of another active surveillance program, this one screening all patients admitted to the adult medical and surgical ICUs of a facility, the overall rate of MRSA infections decreased from 6.1 infections per 1,000 days to 4.1 infections per 1,000 days.13 Of most importance to note concerning this particular study, is that the researchers write, MRSA would not have been detected in 91 percent of these patients if screening had not been performed.

As far as cost, they note that it took $3,475 per month to finance the program, but we averted a mean of 2.5 MRSA infections per month for the ICUs combined avoiding $19,714 per month in excess cost in the ICUs.

Universal Surveillance

Last October a study was released by researchers at Evanston Northwestern Healthcare that demonstrates universal surveillance is a far more effective program than passive or targeted active surveillance when monitoring for MRSA.14 Additional data from the study signals that universal admission surveillance is cost effective and can have a significant impact on reducing nosocomial-related BSIs.

Evanston has been using universal surveillance for MRSA since the summer of 2005 and has since yielded positive results in 5.4 percent of all admissions (1,309 of 24,045 total patients). The researchers note that ICU-based targeted active surveillance correctly identified 478 MRSA carriers, or 37 percent of those detected by the universal surveillance practices. With passive surveillance, currently what the majority of U.S. hospitals practice, only 247 patients would have been identified as MRSA carriers, missing fully 80 percent of the MRSA reservoir for spread, according to an ENH press release. Lance R. Peterson, MD, director of microbiology and infectious disease research at Evanston Northwestern Healthcare, and one of the authors of the study, was quoted in the press release saying, Active MRSA surveillance at patient admittance is critical to reduce the ongoing spread of MRSA within U.S. hospitals.

Peterson specifically mentions that MRSA surveillance is of higher importance in hospitals where more ICU beds exist. Brigham and Womens Hospital is an 800-bed hospital with eight ICUs. They conducted a retrospective study of four major IC interventions using an interrupted time series design to evaluate their impact on MRSA-related infections. Interventions were introduced one at a time during a nine-year period and involved the institution of routine nares surveillance cultures for MRSA in all ICUs upon patient admission to ICU and weekly thereafter throughout the patients stay in the ICU. Positive cultures resulted in the initiation of contact isolation precautions.15

Routine surveillance cultures and subsequent contact isolation precautions resulted in substantial reductions in MRSA infections in the ICUs as well as the non-ICU areas. Within 16 months, MRSA infection incidence decreased by 75 percent in the ICUs and by 40 percent in the non-ICU areas leading to a 67 percent hospital-wide reduction. It is interesting to note that no similar decrease was attributable to any of the other IC interventions studied (i.e. enhanced hand hygiene protocols, etc.), in this model.

The Real-Life Challenges

One great argument against the use of universal surveillance is that of the required manpower it would necessitate. The State of Virginia studied this in a similar manner when, in response to the consideration of mandated nosocomial infection data reporting, a statewide analysis was conducted. The analysis found that an additional 160 ICPs would be required to handle the workload for the mandated data reporting. What is interesting to note from this analysis, is that only 64 percent of the Virginia hospitals had a full-time ICP employed, according to the survey results.

Daniel J. Diekema, MD, associate hospital epidemiologist at the University of Iowa Hospitals and Clinics, says he thinks that active surveillance can be a valuable tool to control MDROs but only in certain settings. I think (active surveillance can be a valuable tool) in a setting of an outbreak or in the setting of hyper endemic rates of infection where your standard or first line control methods are not effective. But I am strongly opposed to a mandated use or universal use of active surveillance cultures because I think that most hospitals are not prepared to safely implement such an approach.

Diekema says that one of the great limitations of the active surveillance literature, and one of the under-appreciated problems associated with active surveillance, are the unintended adverse consequences of a broad implementation of an active surveillance program. This includes consequences in the resource-intensive nature of the intervention, he says.

Despite our best intentions, isolation of patients for infection control does have adverse, unintended consequences, he warns.

Diekema, and co-author Michael Edmond, MD, MPH, MPA, hospital epidemiologist and medical director of performance improvement at Virginia Commonwealth University (VCU) Medical Center, published an article in the April 15 issue of Clinical Infectious Diseases on this very topic.4

 Because I work in an urban safety net hospital, I have been very concerned about this, asserts Edmond. There are so many issues with this whole active surveillance cultures and contact precautions campaign that is problematic. The science is not there. There is no definitive evidence in the literature that this is an effective way to control MRSA. In fact, the only multi-centered randomized controlled trial to evaluate the effectiveness of this showed no difference. (Edmond refers to a presentation at the 17th annual SHEA conference held in April by Mayo Clinic expert W. Charles Huskins, MD, MSc. The presentation is titled New Insights into MRSA: Screening and Reporting: Evaluating Active Surveillance/ Contact Precautions for MRSA and VRE: The Results of the Strategies to Reduce Transmission of Antimicrobial Resistant Bacteria in Intensive Care Units (STAR*ICU) Trial.) 

Edmond continues, When people talk about the studies and the literature, these are mostly quasiexperimental before and after studies that have many methodologic problems most of which did not use the appropriate statistical analysis to even determine whether they had an effect or not.

On top of that, the people that have been driving this are basically zealous. They dont ever emphasize any of the downsides to this. They refuse to recognize that there are major problems with this whole effort.

Diekema and Edmond point to a long list of adverse outcomes associated with isolation in the healthcare environment. Edmond states that there are studies in the literature that document up to a 50 percent reduction in visits to these patients by doctors and nurses, for example.

One of the limitations of existing literature concerning active surveillance cultures and contact precautions is that it focuses too narrowly on infection-related outcomes and ignores patients overall experience of care, the article reads.4 Noninfectious adverse outcomes should be measured and reported, and stratified according to contact precautions status.

Diekema says he believes an active surveillance program can be implemented safely, but only if you attend to those things, he offers. All of those measures you have to be willing to measure as well. We all have the same goal, and thats patient safety. We have to be ready to define patient safety in a holistic way that includes not just if someone gets colonized by a MDRO, but that the outcome of their hospitalization is successful and satisfying to that patient.

Other drawbacks and challenges relate directly to the isolation protocol needed for the success of a generalized surveillance program. Edmond speaks of the already challenging overcrowding within the inner-city hospitals, especially in the emergency rooms (ER). Active surveillance culturing adds to that complexity through its requirement of isolation practices, he says.

It just really complicates it, particularly in hospitals like ours where the minority of beds are actually private rooms. We have already the problem of ER overcrowding and the inability to get patients out of the ER and up to the floor; the inability to get patients out of ICU and onto the floor What studies have shown is that when you start a program of active surveillance for these organisms, you can expect to quadruple the number of patients that need these kinds of isolation precautions. It doesnt take a genius to figure out that if you are already short on beds, adding another level of complexity with a large increase in the number of patients who are going to need isolation precautions, you are going to have even more problems with throughput.

Diekema says a certain amount of molding of the program is essential to any measure of success. What I would advocate is flexibility to allow each hospital to take the approach that best fits their needs, he offers, adding the entire plan requires a touch of common sense. As an example, if a hospital has a 40 percent rate of adherence to hand hygiene and a 25 percent adherence to contact precautions, it makes absolutely zero sense to seek out more patients to put in contact isolation. 

Lions and Tigers and Bears

Another presentation at the SHEA annual meeting pointed to a successful reduction of both MRSA and MSSA (methicillin-susceptible Staphylococcus aureus) infections using a broad-based approach and one that doesnt focus on just one bug.

Edmond says this is the perfect example of a grossly overlooked aspect of surveillance programs. We really should focus on reducing all infections, not just MRSA. Thats only one organism. A lot of hospitals are putting so much effort and money into trying to avoid an infection due to this one particular organism. Why not focus on interventions that actually are much cheaper, have much better evidence in the literature, and avoid all infections, he asks.

Diekema says that is the root of the philosophical concern that he has with active surveillance. If you are having an outbreak with a specific organism like MRSA or VRE, obviously active surveillance can be key to controlling that outbreak. Some hospitals may have continuous outbreaks and they need to apply active surveillance. But as a general approach to infection control, I prefer to focus on evidence-based measures that we know will reduce infections due to all organisms things that relate to reducing all infection-related morbidity not just focusing on a single organism.

He adds that in many of these surveillance-type studies, MRSA rates are indeed going down, but nothing is happing to reduce MSSA.

I dont think the conclusion you want to draw from that is that that was necessarily a great success. Its no great shakes to tell me, Look we reduced MRSA while our MSSA rates stayed the same or increased. That doesnt make sense to me to take that approach. Your goal should be to bring to zero all of your infections.

Reeling It In

Some hospitals have successfully reduced MRSA without using active surveillance. Edmond says that at VCU they apply a very targeted program where they simply use active surveillance only in the burn unit and bone marrow transplant settings of the hospital unless there is some sort of outbreak.

We have not done active surveillance, but have been able to reduce our MRSA infections, he notes. In fact, our reduction of infections are actually better than some of the hospitals actually doing active surveillance. In our ICUs, we have reduced our MRSA infections by over 70 percent.

He says the VCU model targets simple interventions such as good hand hygiene, avoidance of lines placed in the femoral regional, very strict adherence to good infection control practices when placing central lines, mandatory education for all house officers when placing direct lines, and giving feedback to the frontline healthcare workers on adherence to each of the measures by quarterly postings on the ward.

Its been very, very successful, he says.

Diekema shares that he has no quarrel with a hospital that wants to use active surveillance. My concern is if a hospital is forced to do active surveillance, they might end up pouring a lot of resources into that intervention and if they arent attending to all the other issues, they could end up spending a lot of money and then not having a lot to show for it.

The key to making any of this successful is to get all of the right players to the table from the very beginning, and get strong hospital administrative support, he advises. It is an intervention that has a broad impact on the facility everywhere from the laboratory to bed management to your admitting physicians to your interactions with long term care and rehab facilities that you discharge to. It is such a complex intervention.

Edmond takes the conservative approach to this topic and notes that active surveillance should be your option of last resort to control MDROs.

If you have tried everything else, and you still cant control MDROs in your facility, then do it, but short of that, I think you have many other things that you can do that will give you much bigger bang for your buck and prevent many more different types of infection than just doing active surveillance for MRSA.