Robert A. Weinstein, MD: Hello, I am Dr. Robert Weinstein, Professor of Medicine at Rush University Medical Center and Interim Chairman, Department of Medicine at the Stroger Hospital of Cook County and Chief Operating Officer at the Ruth M. Rothstein CORE Center, all in Chicago. I would like to welcome you today to this MedscapeCME spotlight on successes in hospital-acquired infection prevention research over the past decade. I am pleased to be joined today by my colleagues Dr. John Jernigan, Deputy Chief, Prevention and Response Branch Division of Healthcare Quality and Promotion at the Centers for Disease Control and Prevention here in Atlanta, Georgia; Russ Olmsted, epidemiologist in Infection Prevention and Control Services at St. Joseph Mercy Health System in Ann Arbor, Michigan; and Dr. Ebbing Lautenbach, Associate Professor of Medicine and Epidemiology in the Infectious Disease Division at the University of Pennsylvania School of Medicine in Philadelphia, Pennsylvania. I welcome you all to today's round table discussion.

The participant learning objectives are to:

• Identify areas of research in the area of hospital-acquired infections (HAI) that led to important clinical advances in the past decade;
• Explain how the findings of current HAI research initiatives will potentially impact clinical practice; and
• Recognize important clinical questions that are currently not being sufficiently investigated.

Dr. Jernigan, major achievements in infection prevention over the past decade have taken place. What is the road map that has gotten us from the bench to the bedside in some of these advancements?

John A. Jernigan, MD, MS: I would like to address that first by talking about some of the research advances that have had a major impact on infection prevention practice today, and as you know, there have been many over the past decade. I have chosen 3 to highlight that I think illustrate the impact of the research we have. First is the use of alcohol-based hand rubs in clinical practice. Before the past decade, the major method of decontamination of the hands was the use of soap and water.

The limitations of this procedure included that it took a long time to do, optimal adherence depended on the number of sinks and the location of sinks in the hospital, and the use of detergents can be very irritating to the skin. All of these can add up to barriers to adherence. And we know that has been a big problem.

Many studies have shown that adherence and hygiene among healthcare workers was very low. But then, alcohol-based hand rubs became available and there was great promise because there was the potential for a decrease in the amount of time that it took to decontaminate the hands. Their [alcohol-based hand rubs] use was not dependent on the location and number of sinks, and it fit more naturally into the flow of work at the patient's bedside.

Research played a very important role, first of all in documenting in numerous studies that these hand rubs were more efficacious than soap and water for reducing the number of bacteria on the hands of healthcare personnel, and also in several studies, alcohol-based hand rubs containing emollients were shown to be less irritating to the skin than soaps or detergents that are often used.

In 2002, the Centers for Disease Control and Prevention [CDC] through the Healthcare and Infection Control Practices Advisory Committee published guidelines for hand hygiene in healthcare settings and firmly established alcohol-based hand rubs at the center of hand hygiene practices, recommending them for routine decontamination of hands in all clinical situations except when the hands are visibly soiled.^[1] This had a major impact on the way hand hygiene is performed in the United States.

A 2008 study showed that 84% of US hospitals surveyed indicated that they had adopted alcohol-based hand rub.^[2] A number of studies have shown dramatic increases in adherence to hand hygiene and it is believed that the use of alcohol-based hand rubs has been a major factor in that success.

Another important research advance is the preventability of central line-associated blood stream infections (CLABSI). Although recommendations for preventing CLABSI were available [in the early 2000s], reports of successful prevention were limited and were mostly single-center reports. However, in the 2000s, there were 2 separate reports of large collaborative regional demonstration projects that focused on improved implementation of existing recommendations to prevent CLABSI among patients in intensive care units (ICUs), first in southwestern Pennsylvania in 2005 and then in Michigan in 2006. Both of these studies demonstrated remarkably similar results, showing approximately 70% reductions in endemic CLABSI rates across a wide variety of facility and ICU types, suggesting that the preventable fraction of these infections was perhaps much larger than we had originally thought.

The results of these 2 studies have contributed in a very substantial way to a major change in the expectations of CLABSI prevention programs. The earlier single-center reports were viewed by many as somehow aberrant or the result of special circumstances or resources that existed in those particular reporting facilities. These regional studies demonstrated that better implementation of existing recommendations can have a major impact across a wide spectrum of US hospitals. Dramatic success was possible, and not just under special circumstances.

These types of programs have now become widespread in ICUs across the United States and there is evidence that nationally, CLABSI rates have declined significantly in recent years, based on data from the National Healthcare Safety Network. These important studies have led to high hopes that improved implementation science, that is, improving implementation of existing prevention practices, is likely to have major impact across the entire spectrum of HAI.

Another innovative advance that takes a different approach is the role of source control in preventing infection, particularly with the use of chlorhexidine bathing of patients. Based on data suggesting that colonization of a patient's skin is an important source of spread for certain epidemiologic imported bacteria, you and your colleagues, hypothesized that daily bathing with a skin antiseptic -- in this case chlorhexidine gluconate -- would decrease the burden of the patient's skin contamination and indirectly decrease contamination in the environment and of healthcare worker's hands, and potentially play a role in decreased transmission of pathogens.^[3]

And although there is still more to be learned about this particular strategy, data are strongly suggestive that daily chlorhexidine bathing can significantly reduce contamination of the patient's skin, the environment, and healthcare workers' hands, and an impact on methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococcus (VRE) acquisition has been documented.^[4]

Several studies have suggested a favorable impact on primary bloodstream infection rates. This approach seems to have had an immediate impact on practice and reports suggest that many hospitals in the country are now using this approach.^[5-7] Again I caution that we have more to learn about this strategy.

Two randomized, controlled trials are in progress that are examining this issue that stemmed from the CDC Prevention Epicenter Program, and we anxiously await the results of these studies.

These are just 3 prominent examples from among the many that exist from the past decade, and I would like to take a step back and examine briefly what is behind those successes because it is important as we think about HAI prevention research agendas for the future. These successes were the result of an entire portfolio of individual research accomplishments, with each successive and progressive accomplishment built upon the observations of previous work.

One can envision a pyramid and at the tip of this pyramid is the ultimate goal -- that is, improving health outcomes by preventing HAI. The tip of the pyramid is built upon a solid foundation that represents several steps in translating basic epidemiologic evidence into successful prevention.

Basic Research: At the bottom of the pyramid we have what epidemiologists call our basic research, which is observations made through surveillance, and outbreak investigation of epidemiologic studies.

• Translating -- The first step in translation is taking that evidence and hypothesizing candid interventions that might have benefit and testing these for the first time in real patient population and healthcare settings.
• Study in Larger Populations -- Only a few of those candid interventions will likely show promise, so the next step in the translation is to take these promising candid interventions and study them further in larger patient populations and in different clinical settings to establish firmly their efficacy and effectiveness, so that they can end up in evidence-based guidelines.
• Incorporate Into Guidelines -- But as we all know and we have seen from the past decade that being present in evidence-based guidelines doesn't mean that it is actually going to be implemented in practice. So there is a whole emerging science of diffusion and dissemination and delivery research that helps translate evidence-based guidance into actual practice at the patient's bedside and hopefully the result, again at the top of the pyramid, is real-world outcomes of improved health.
• Improved Health -- We need to continue to monitor those because patterns and trends that we see there can help inform back at the bottom of the pyramid new hypothesis generation and new studies.

Each of these phases is important and I think that future prevention successes depend on investment and attention in each phase of the research process.

An example of where this can break down that I told you about is the success in CLABSI prevention. It took many, many years from the existence of evidence-based guidelines to realizing the public health benefit of implementing those [guidelines]. The link that was missing was this diffusion and dissemination and delivery implementation research. If we were able to pay more attention to that, we might have been able to expedite that process.

And so, in the future, we need to pay attention to each of those steps in the translational process. We have seen some great successes in HAI prevention research over the past decade and these were built upon the foundation of a successful research pipeline, if you will. We are not finished learning yet. We need to nurture and support HAI prevention and research to improve upon existing recommendations and to develop new recommendations. And as we look to the promise of the future of HAI prevention research and as we develop HAI prevention research policy, research infrastructures, and research agendas, it may be helpful to consider the translational research framework to guide us. To get closer to our theoretical goal of eliminating all HAIs, we have to remember each of the steps involved in translating basic epidemiologic evidence and successful prevention strategies, and devote adequate attention and resources to every step of that research process.

Dr. Weinstein: Thank you. Russ, we have heard how researchers have translated their advances from basic concepts to the bedside, sometimes faster and sometimes slower. What do you see as some of our home runs and how generalizable are they?

Russell N. Olmsted, MPH: Great question Bob. To echo what John mentioned, the prevention and almost elimination of CLABSI has been a demonstrable home run, and I think we have reached the tip of that pyramid, as it were. That site of infection has lent itself very well to elimination or major progress and, primarily building on foundational research.

The general term used now is "bundling" of various prevention practices. Lots of recommendations that came out of the basic research that John mentioned have been put into just an organization [the central line bundle] of a number of things that happen at the time this device is inserted. This has had a dramatic impact, as John mentioned, an almost two-third reduction in the frequency or incidence of CLABSI. So I think that is clearly a home run, and some of the early building blocks or foundational research came from studies done by folks at Johns Hopkins University. Dr. Berenholtz and colleagues looked at taking these elements from CDC prevention guidelines, organizing them into a composite of approaches, and then applying them at the bedside.^[8] As a result they had a reduction of the baseline frequency of bloodstream infections of over 11/1000 central line days to basically 0/1000 central line days, a dramatic reduction at 1 institution.

This had been built on and expanded further, kind of moving up that pyramid, where we expanded this to a number of centers, hospitals, within a state. The experience that I have had recently in Michigan was a project led by Dr. Pronovost from Johns Hopkins as well, expanding what they did initially at their facility and seeing whether it would work in the real world with a lot of facilities.^[9] Well over 76 hospitals and 125 ICUs were enrolled in this bundle project looking at CLABSI prevention, and the results were similar to what happened at the single facility, again a two-thirds reduction in the frequency or incidence of CLABSI.

The bundle consisted of very simple procedures, which were well-outlined in CDC guidelines, including hand hygiene by the person inserting the device, maximal barrier precautions, chlorhexidine gluconate for antisepsis applied to the site of the insertion, avoidance of femoral central line insertion, and removal of the central line as soon as possible or take it out when it is no longer needed. Those 5 procedures really had a dramatic, sustained impact throughout these centers in Michigan. There has been a follow-up study looking at ways of maintaining this gain, which has been held up to this time.

Another concern that John mentioned is that we still have challenges. When you move outside of the ICU where we have an organized, intensivist-led ICU, nursing-led team that works very closely together, we have issues because we don't have as much organization around that. There might not be a single lead physician or nurse. We have multiple attending physicians with different patients that may enter that facility. There are at least 4 studies that I am aware of published since about 2003 looking at the problem of bloodstream infections that now seem to have migrated out of the ICU into the standard medical-surgical arenas.^[4,10-12] Our challenge now is to take that bundle approach and see if we can apply it in that setting, which is kind of the new frontier.

Many of us are familiar with the CDC surveillance system, originally the National Nosocomial Infection Surveillance System (NNIS) and now the National Healthcare Safety Network (NHSN). That in my mind is also a collaborative, as it were. There was an organization of hospitals that were initially concerned with HAI. They wanted to collect data to see whether they were making an impact, so they organized a very significant collaborative for standardizing methods for surveillance of those types of approaches. As a result, we have seen a dramatic reduction in participants within the NNIS and NHSN. Again, almost a two-third reduction in CLABSI. So, this building block, the pyramid of evidence, really has helped up to the current time.^[13]

The experience in the Pittsburg Regional Healthcare Initiative was also a dramatic reduction, a two-thirds reduction in CLABSI.^[14] They used similar processes to the Michigan collaborative, the same elements of the organized bundle approach.

Dr. Weinstein: So 5 simple things down to 0.

Mr. Olmsted: Yes, exactly. Those 5 easy steps. Another challenge that folks who take care of central lines have realized is that once the device has been placed and is in place for a number of days, we get into care and maintenance issues that we don't have as much information on in terms of an organized bundle. That will potentially come in the future.

Other evidence of the collaborative; again, good evidence that this organized approach works not only at a hospital, but also statewide at several hospitals within a region, is the Duke Infection Prevention and Control Network.^[15] They have looked at a number of initiatives, bloodstream infections being one.

They also looked at incidence of colonization with MRSA, ventilator-associated pneumonia (VAP), and occupational injuries of personnel while they are taking care of patients. The incidence of all was reduced with the implementation of evidence-based tools for effective prevention of transmission of infection at the bedside. It is not just with CLABSI that we are seeing results.

Additional evidence has recently been provided by Bouadma and colleagues,^[16] also a collaborative approach, which showed a greater than 50% reduction in VAP. From that study and nationwide we have had good evidence from the Surgical Infection Prevention and Surgical Care Improvement Projects led by the Center for Medicare and Medicaid Services and Dale Bratzler's group for dramatic reduction in surgical site infections (SSI).^[17] Finally, Wald and colleagues^[18] looked at catheter-associated UTIs [urinary tract infections] -- morbidity and mortality associated with the device -- and trying to get the catheter out by postoperative day 2. There is good evidence that this organized approach makes a real difference.

Then most recently Haynes and colleagues^[19] looked at 8 major centers throughout the world. They organized a checklist of things to go through before surgery is started.

There was a dramatic reduction in SSI and other postoperative complications.^[19] I think we are clearly in the era of collaborative studies and this approach has had a major impact in terms of preventing these complications.

Dr. Weinstein: So a number of home runs related to devices and procedures. Ebbing, you have had a long-standing interest in the investigation of antibiotic resistance. Because we haven't talked about the bugs yet, what do you see as current issues in multidrug-resistant organisms (MDRO) and have we been as successful with those as we have been with some of these devices and procedures?

Ebbing Lautenbach, MD, MPH, MSCE: That's a great question. As John and Russ have mentioned, there have been a lot of successes over the course of the past 10 years. Certainly, a lot of challenges remain and there are a lot of questions that we still have left to answer. The issues not only of HAI in general, but also particularly HAI due to antibiotic-resistant organisms, remain a big area for future work. I think the goal for us is to figure out moving forward how we build on the successes that we have had to date and use that as a foundation to move forward. This is a great time to discuss this because the current climate is probably better than it has ever been in terms of generating interest and potential support for conducting studies that look at HAI and particularly resistant infections.

The federal government including funding agencies, the Institute of Medicine, the Joint Commission, scientific societies, the lay press, and even the general population are probably more tuned into the issues of HAI and antibiotic resistance than they have been previously. This is a great time to really think about what those next steps are because the current climate lends itself very well to us being able to move forward. There are a couple of areas that I think are really important. That is not to say the ones that I talk about are at all an exhaustive list. There are certainly plenty of things to talk about, but I think there are a couple of areas in particular that really require highlighting.

One is MRSA. We have done a lot of work in MRSA over the course of the past 10, even 20 years, trying to figure out how best to prevent the further emergence of MRSA. One area of controversy is the role of active surveillance, or what the value is of actively screening patients for MRSA. This is a very controversial topic, with a lot of different viewpoints, and even the different societies have come to different conclusions and recommendations. Coupling that with the recognition that there are a number of states in the United States that have mandated some form of active surveillance for MRSA, makes this a topic that we really need to pay more attention to than we have in the past.

Most of the studies done in the past, and I think this is where the controversy comes from, were typically small, single-institution studies, and often with quasi-experimental, pre-post design. The results from those studies leave the conclusions open to interpretation and raise the issues of potential confounding or bias. With that said, even more rigorously done studies including 1 from Switzerland^[20] and 1 from the United States^[21] conducted more recently, have come to different conclusions. So clearly, more work needs to be done.

Russ talked about the importance of collaborative enterprises and multicenter studies and this is one area where we can really benefit from large multi-center studies looking at different approaches to curbing MRSA, and particularly standard approaches across institutions to give us the best data that we would need to better inform what we do in the future.

John, one of the things that you mentioned was implementation science, so studying exactly how these sorts of strategies are even rolled out [is important]. We have very little information about what the best approaches are for rolling out these sorts of interventions, even if we knew what the right interventions are.

A lot of work has been done on MRSA and gram-positive pathogens, in general. One of the areas where very little work has been done is gram-negative pathogens, such as Klebsiella, Acinetobacter, and Pseudomonas. Even though we have spent a lot of time on MRSA, it is well recognized that MRSA accounts for only 8%-10% of HAI, so we are missing a lot of potential pathogens by simply focusing on the gram-positive pathogens. There are some data that suggest that gram-negative pathogens are becoming more common as a proportion of all HAI.

One of the real concerns here is that many of these organisms are now multi-drug resistant to the point where there are sometimes few and occasionally no antibiotics that we have to choose from. And the real concern is that if current trends continue, we are going to end up in what has been called or could be considered the "preantibiotic" era, returning to a situation where we have patients, particularly in healthcare settings, for whom we have absolutely no agents to treat their infections. Moreover, one of the things that we have noticed over the course of the past 10 years is that what antimicrobial drug development has occurred, has focused on drugs that target more expanded gram-positive coverage, not gram-negative coverage. And so, dealing with resistance -- meaning developing new drugs -- is clearly not happening for gram-negative pathogens.

Part of the issue is that we don't really know what is driving gram-negative resistance. How much of it is how we use antibiotics, perhaps overuse them, and use them inappropriately vs how much of it is person-to-person spread? If we knew more about those particulars, we might be able to better target strategies for figuring out what to do about gram-negative organisms.

Antibiotic stewardship has gotten a fair amount of attention; these are programs that are geared toward helping clinicians, particularly in healthcare settings, to figure out how to best use antibiotics and which antibiotics to use in different settings.

Even though there are a fair number of programs now in existence (ie, guidelines from the IDSA [Infectious Diseases Society of America]^[22]), we still don't have a whole lot of data about what are the most effective components of antibiotic stewardship programs, and so we clearly need more work in that realm as well.

I have spent a lot of time talking about the hospital setting and the reason for that is because that is where we have most of our data from. We tried to extrapolate data from the hospital setting to other [nonhospital] settings, such as long-term care, rehabilitation facilities, etc, although we know that these settings are very different from the in-hospital setting. We clearly need to do a lot more work to figure out what is going on not only with infections in general, but particularly with antibiotic-resistant infections, in these other healthcare settings, including long-term care, rehabilitation, same-day surgery, and even home care as more and more healthcare services move to settings outside the hospital.

Despite the vast bulk of the data that we have from the hospital setting, we still don't know what interventions, potentially antibiotic stewardship interventions and infection control interventions, might work in these settings. And there is good reason to think that they won't work terribly well. For example, a long-term care facility is where people reside, and the same sorts of things that we do for infection control in the hospital are clearly not going to work in those settings.

And one thing to recognize is that while we may focus on just the hospital or just the long-term care setting, many times patients move from the long-term care setting to the hospital, back to the long-term care center or a rehabilitation facility. Knowing more about transitions of care and how we can better identify patients who are likely to become infected or colonized with resistant organisms, is going to help each of those particular entities to solve the problems of infections and resistance. Overall, there have been a lot of successes as John and Russ have highlighted. There are clearly a lot of other questions that remain to be answered and this is a great time to be thinking strategically about how to move forward.

Dr. Weinstein: Thanks Russ. You talked about 5 simple things to get CLABSIs down. And in the past I have thought that if you could either improve asepsis, that is get people to behave better, or develop a fool-proof device, you are better off with the device. But you didn't mention devices at all. Has the collaborative experience shown that I am totally wrong?

Mr. Olmsted: No, I was anticipating your question so it was perfect. There are lots of tools and techniques available. Our experience with these collaboratives has been that because of the reductions that have been achieved without the use of the newer technologies, such as antimicrobial central venous catheters, we have looked at whether we can achieve this reduction without necessarily going to the additional cost of the device. The antimicrobial catheters cost at least some fraction more than a standard catheter. Based on the success that we have seen with these collaborative, my advice is to try those 5 steps first, look at care and maintenance issues, and then if there are still issues or the rates are still high, then we can look to devices that might help us prevent those infections.

Dr. Weinstein: What I like about it is that it changes the culture rather than just bringing in a quick fix. So John, if I were the chief executive officer and I came to you and I said, there are 5 things here. If we do only 3 of them maybe it will cost us less, which 3 should we pick? How do we deconstruct these bundles and figure that out, or can we, or should we?

Dr. Jernigan: Well it is tough. To figure out the incremental contribution of each element of a bundle can be a difficult thing to do. It requires the cooperation of a large number of centers, a very organized study, and to be honest with you, we need to ask ourselves, is the juice worth the squeeze? So in some cases for elements of the bundle that fit straightforwardly into the flow of work that aren't horribly expensive, maybe we just accept that these 3 things or these 4 things go together and we don't try to tease out the incremental contribution of one or the other. For some bundles there may be a part that is particularly expensive or particularly difficult to implement. One example is bundles for preventing MDRO and including active surveillance as part of that bundle has been suggested. That is a very controversial issue and I think for that particular piece, we do need further study it to decide whether that is a critical component of the bundle. So the answer is that it depends, as always.

Dr. Lautenbach: In the past we have been very focused on answering these very precise scientific questions. What happens if we do this 1 intervention and what does that do to the outcome? In some ways that has been a stumbling block because what has happened is that we have a number of small studies that have each looked at a couple of different interventions, rather than looking at what makes sense as a bundle of interventions. And in some ways that has been a bit of an obstacle to thinking intuitively about the things that best fit together that might limit a particular type of infection.

If you find that a bundle works, and it is a scientifically interesting enough question or there is a compelling reason as John mentioned, to go back and tease out what specific parts of that might make sense, I think that is a reasonable approach. But, I think that evaluating the bundle as a whole increasingly makes sense.

Dr. Weinstein: So you guys are not up for deconstructing.

Dr. Weinstein: UTI, we haven't talked much about those. They are not a huge part of morbidity and mortality and not a huge cost, but they are a problem and if you know anyone who has ever had a UTI, they are very unpleasant. And there are a lot of resistant bugs in the urine. What are some of the things that can we do, Russ? What is the bundle for UTI prevention?

Mr. Olmsted: We call it CAUTI (catheter-associated UTI) prevention or the bladder bundle, if you prefer. Or the other term is the Dangerfield project because it is a site that doesn't get respect. That was my best material for the day. With CAUTI it has been more about looking at processes around the insertion of the device. Can we minimize the overuse of the catheter, and thereby ultimately prevent infection, or at least starting toward that pathway of a bacteriuria or a true CAUTI.

Most of the big projects that I'm aware of right now have emphasized processes of care to make sure that you identify the patient who needs the urinary catheter, to ensure that it is clinically indicated. Once you establish that or put the device in, get it out as soon as possible. Most of the metrics on CAUTI that I have seen have not looked at the outcome of UTI as much as at the proportion of catheters being used appropriately.

That seems to be the emphasis with CAUTI. The data are much easier to retrieve from the patient bedside.

Dr. Weinstein: John, you know I always quote that 60% of bladder catheters are not needed, and if there is no bladder catheter, there is no CAUTI. How successful are we in getting catheters out when we don't think they belong there? But first, how do we know they don't belong there? How do we decide that they don't belong there? How successful have we been in removing them?

Dr. Jernigan: There is a new CDC guideline for CAUTI prevention that has a heavy emphasis on appropriate use of the urinary catheter, and I agree that the mainstay of CAUTI prevention is not using the catheter when you don't need to and I would encourage our viewers to look at that guideline ^[23] because it outlines specific criteria to help you decide who needs a catheter and who doesn't.

The answer to the question of how successful we are at getting the catheters out, I think the general answer would be "not very" at this point. And it probably varies a lot as well by setting. We might be doing a little bit better in the acute care setting. When we move to the long-term care setting, that is a particularly problematic area for a number of reasons and research can play an important role there. For example, you often hear of people using urinary catheters to protect patients who may have sacral decubiti or perineal wounds, and there are some that would even argue that urinary incontinence is a risk factor or makes those problems worse. We really don't know the answer to that and we need studies to suggest what the role of urinary catheterization or urinary drainage is in either preventing or treating those particular problems. We need to know more about the use of external drainage systems compared with indwelling bladder catheters. There are a number of research questions there that would help us answer that question and get closer to the truth.

Dr. Lautenbach: Even if we can't identify in what situations a catheter is needed, there are plenty of data to suggest that even the attending physician for a given patient in the hospital setting rarely knows whether the patient has a catheter in place. And so, this is low-hanging fruit that we can target -- simply having the attending every day figure out whether the patient has a catheter and whether they really need that catheter. That could be a case-by-case decision guided by the guidelines. Again, that seems to be very low-hanging fruit that we can target to make inroads into reducing UTIs.

Dr. Jernigan: We may be able to make use of the emergence of electronic medical record (EMR) systems. There is some work going on in the CDC Prevention Epicenter Program to see if they can use data available in the EMR to create reminders to physicians that (a) this person has a catheter, and (b) to ask several questions to help them evaluate does it really need to be there today or can it be removed.

Dr. Lautenbach: I think John's point that we don't have much if any data in the long-term care facility setting is an important one because, while UTIs may be the ones who don't get any respect in the hospital setting, that is really the main thing that we are talking about in the long-term care facilities and we have very few data that suggest when people need a catheter, when do they not, and how to best evaluate that. So I think that is clearly an area for future research.

Mr. Olmstead: On that theme of antimicrobial resistance or MDRO, I'm curious about your work, Bob. Do you have any sense of the capability or the efficacy of patient bathing for some of the gram-negative pathogens that Ebb was talking about?

Dr. Weinstein: Yes, we have seen and others have reported control of Acinetobacter outbreaks with chlorhexidine bathing, so there seems to be a benefit, because the gram-negative organisms on the skin are a problem. If the gram-negative organism is in the gut or in the urine and that is how it spreads out to the hands -- through the gut or urine -- [chlorhexidine] bathing is not going to make a difference. But Acinetobacter, which is on the skin a lot, is a great target for chlorhexidine.

Dr. Weinstein: You mentioned something very interesting, the EMR, a big part of the government push is to get everybody on EMRs. What do you think Ebb? You've had a lot of experience with this at Penn.

Dr. Lautenbach: There are a lot of situations for which the EMR, and especially decision support, can play a huge role in improving how we provide care. That goes from reminders of drug-drug interactions all the way to decision support where you put in, "I have a patient in whom I suspect sepsis and here are the antibiotics that they have gotten recently" and it provides recommendations for antibiotics.

That goes a long way toward replacing or at least optimizing what we do currently for antibiotic stewardship. What we have found in the past 10 years -- and there has been a real push toward the use of the EMR and computerized decision support -- is that there are a lot more problems than people originally anticipated.

You can put up all of the alerts that you want to assess whether that catheter still needs to be there, and ask clinicians if they're sure the patient still needs this antibiotic, but there is clearly alert fatigue. If you have all of these things that pop up on the computer screen, the people who put in the orders get very, very savvy at figuring out how to work around those alerts, so we need to be careful about how we do that and specifically target those things that are most important. That's not to say that there isn't a lot of promise, but there are a lot more obstacles than we initially anticipated when this was first proposed as a sort of cure all for many of the things that happened.

Dr. Jernigan: Another advantage of the emergence of the EMR has to do with measurement and monitoring. You know a lot of what our infection control personnel do now is spend much of their day counting infections, counting events, and counting denominators, and that takes a lot of time and it takes away from the actual promotion of prevention.

A very fruitful and exciting area of research is using data from these electronic medical record systems to create surveillance algorithms that can be simplified or even completely automated that provide several advantages. First, it frees up a lot of time for infection control personnel to focus on prevention rather than bean counting. And second, there is evidence that suggests that it is a better way to measure because you take the human element of interpretation out of it. By developing very objective parameters it might be a much more useful metric for determining the impact of your interventions.

Dr. Weinstein: Russ we haven't talked about Clostridium difficile, antibiotic-associated diarrhea, a huge problem with a big increase in cases in the United States, and a bad strain emerging. What are you guys doing in Michigan? Are you seeing a problem? How do you deal with it?

Mr. Olmsted: Yes, it is definitely a problem as in other states, a dramatic increase. About 2000 we saw a nationwide uptick in the frequency of C difficile infection and concomitant with that was this new strain that seems to produce more toxin or cause more virulent infection. We have had some studies with the support from the CDC, did some investigations of molecular typing of the strains, and found 1 cluster; in 1 patient unit we documented 3 patients with the same strain.

This pathogen is moving around, probably on the hands of healthcare personnel, but I think the bigger issue has been the environment of care, which is a big focal point for trying to improve that. We are emphasizing frequent cleaning and disinfection of surfaces around the patient. Surprisingly, new research has shown that that if you are admitted to a patient room that has been previously occupied by a patient with C difficile infection, your risk for acquisition is higher,^[24] suggesting again that this is probably an environmentally mediated pathogen in terms of the mode of transmission from the focal point.

Dr. Lautenbach:This is another situation in which the emphasis on antibiotic stewardship makes a lot of sense. For decades, antibiotics have been identified as a clear precursor for people acquiring C difficile, making it more likely for them to get that in the hospital. Optimizing antibiotic use within the hospital setting, but also in the other healthcare settings, would go a long way to addressing that.

Dr. Weinstein: If we had a problem at our hospital and we called the CDC and said, Well, bundles are in, so what is the bundle for C difficile? What should we be doing? What would you say?

Dr. Jernigan: That is a good question. A big part of the bundle should be as Ebbing mentioned, appropriate use of antimicrobial drugs, which is a little outside the scope of infection control, but very important in taking care of this pathogen. There is a role for additional precautions, over and above standard precautions, such as the use of contact precautions for patients who have C difficile, and in some cases patients who have diarrhea who you are not sure whether they have C difficile or not. Russ mentioned as well that particular care to the environment for this particular pathogen may have an important role.

I talked about the use of alcohol-based hand rubs and this is an area in which there is a little bit of chink in the armor, because as you know, the alcohol-based hand rubs don't have significant activity against spores and C difficile is a spore-forming organism. We know that spores can play a prominent role in transmission on the hands and in the environment. If you have evidence of ongoing transmission in your hospital, you might want to consider making exceptions to the use of alcohol-based hand rub for those particular patients who you know have C difficile, by switching to soap and water.

Mr. Lautenbach: John Boyce conducted a nice investigation of the correlation between alcohol-based hand rub and incidence of C difficile infection and found that there was no correlation.^[25] Mudo and colleagues^[2] looked at a major outbreak of C difficile infection at their academic center and it actually peaked before alcohol-based hand-rubs were introduced, so I think there is some indirect science. There is certainly laboratory evidence that spores survive very nicely in alcohol for years. To me the real key is personnel. If they are using contact precautions, the key is to put on gloves, minimize the contamination on the hands, and for most endemic situations, make sure you take the gloves off. Use whatever hand hygiene methods you prefer; I am not as convinced that the soap and water has to be an absolute, unless there is a real cluster.

Dr. Jernigan: I agree with that. In fact, our recommendation is that in most cases, routine use of alcohol-based hand rub should remain the standard. But, if you are having a particular problem and can't control transmission of C difficile, you might consider this supplementary measure of going to a different hand hygiene.

Dr. Weinstein: Have they mentioned gloving for C difficile, universal gloving? Should we be putting gloves on and taking them off every time we see a patient in an acute care hospital?

Dr. Lautenbach: That is a great question and unfortunately, one for which there isn't great evidence for or against. One of the things that we have identified is that if you gown and glove when you go into a room, you are more likely to attend; it just makes you think more about contact precautions and exercising good hand hygiene. But, when some patients get gown and gloves and others don't, how good are we about doing that every time that we go into a room with a patient where it should be done? We are not great.

And so, there is some intuitive appeal to having a standard approach, much like standard precautions, only this would be standard gloving whenever you go into someone's room as a way of preventing transmission of any one of a variety of different organisms. The intuitive appeal is that you do the same thing in every room. That is what is expected, you don't have to check signage or slip in quickly so as not to touch the patient because you are only going into the room, and those sorts of excuses that you hear for not gowning and gloving.

Dr. Weinstein: John, what does the CDC say, not to put you on the spot?

Dr. Jernigan: On the use of universal gloving? We don't know. That is a fruitful area for research. I will say this, if you are going to try that approach, you need to make sure that the healthcare workers understand that they need to change their gloves. We have seen and heard in a number of situations where they say, Oh that means I can wear the same pair of gloves the entire shift.

Dr. Weinstein: We have talked a lot about settings outside the ICU. Does the CDC collect any data yet? The NHSN is ramped up and is getting data from more than 1500 hospitals now, I think. Are we getting any data from long-term acute care hospitals or from nursing homes?

Dr. Jernigan: We are working carefully with long-term acute care hospitals to try to share some of their data and we are working actively on building modules in the NHSN that can be used by both long-term care facilities and long-term acute care hospitals. Another outpatient setting in which we hope to have modules available is the ambulatory surgical care setting. We know that the number of surgical procedures being performed in this country in ambulatory surgery centers is exploding; it is huge, and so we need to have a measurement system available to get a handle on what the burden of infection is.

Dr. Weinstein: Something is wrong, go where the money is. I am impressed that the CDC is doing that. Ebb, what can we do to engage nursing homes? So we are going to have these modules, why would they want to use them?

Dr. Lautenbach: I do a fair amount of work in long-term care, and they really are focused on these issues. There has historically been somewhat of a disagreement between hospitals and long-term care facilities as there was a lot of patient movement back and forth and in the early days of MRSA and VRE. Each side would blame the other that this patient obviously got VRE at your place because they didn't have it here. I think a lot of those issues sort of resolved, because, and as I mentioned earlier, of the recognition of the importance of these transitions between care. So, I think long-term care facilities are invested in this. There is a clear recognition that when infections occur in residents of long-term care facilities, specifically those who need to be transferred to the hospital, there are huge morbidity and mortality rates, and they have a vested interest in addressing this.

One of the interesting things as I mentioned is that there are number of state mandates for screening for particular resistant organisms, specifically MRSA, and in some states those mandates extend to long-term care. As I mentioned, this is a unique climate that we are in for addressing some of these issues. This may the first in-road for which we have fairly comprehensive data, at least in as much as it is collected in the same way from place to place, where we can engage not just a nursing home here and a nursing home there, but a whole network of nursing homes and perhaps even state-wide to address some of these issues.

Dr. Weinstein: Gram-positive and gram-negative organisms, which is a bigger problem in nursing homes?

Dr. Lautenbach: They are both big problems. The sad thing is that we don't have great prevalence data for either in long-term care. The data that do exist suggest that both MRSA and resistant gram-negative organisms are greater in number in long-term care than even in the ICU.

Dr. Jernigan: Everyone would agree that prevalence is very high in nursing homes, and we hear the situation a lot where the hospital is blaming the nursing home and the nursing home is blaming the hospital. We really don't know the answer to that question. How much transmission is actually occurring in long-term care facilities vs is that just the place where patients who are at a high risk for carriage of MDRO go and naturally tend to collect? That is an important epidemiologic question. Should we focus our efforts to prevent transmission more on the acute care side or more on the long-term care side? That is unclear and it is a very important question as you know, because many of the measures that we use in the acute care setting to prevent transmission are just not practical in the long-term setting.

Dr. Weinstein: Russ, in your system in Ann Arbor [Michigan], how do the nursing homes and your hospitals get along?

Mr. Olmsted: It is a fairly supportive environment. We have a counsel that meets regularly with local long-term care facilities. We have a hospital within a hospital, long-term acute care. There is a notable increased resistance in the strains that are much more multidrug resistant compared with the strains that we see in our inpatients.

Dr. Weinstein: The gram-negative organisms?

Mr. Olmsted: Yes, predominantly gram-negative organims. Our challenge is to figure out what works well in both of those settings and what will prevent transmission. Long-term acute care is an area that really needs a lot of research right now.

Dr. Weinstein: Something that occupied everybody's time and mind and efforts in the last 6 months was flu, H1N1, swine or novel H1N1, whatever you want to call it. Hype? What did you think about that experience? Did it help us prepare for the next? Was it a lot of spinning of wheels? How did you interpret that?

Dr. Lautenbach: It was a unique experience in that much of what happened after 9-11 and the anthrax attacks of 2001 was gearing up for something like this, and so it was a unique opportunity to figure out how well in fact we are prepared. We did a survey of the Society for Healthcare Epidemiology of America membership, hospital epidemiologists, people who we figure would be on the front lines dealing with this. What was surprising was that upward of about 75%, 80% of people thought that they were pretty well prepared for what happened. This was still in the throes of H1N1, but it had been far enough along that I think had things really not been going well, they would have known at that point.

What was interesting was that although they felt well prepared, they recognized that this was such an overwhelming demand on their time, all of the other things that happened for infection control completely fell by the wayside. I think that is something that, not if but when, this happens again, we need to think about what contingencies are in place to make sure that everything else doesn't get dropped.

Dr. Weinstein: I brought that up at the end because I think it points out that our surveillance systems are good and our communications are getting better. Our idea of bundles and gauging healthcare workers is getting better and so I think we were prepared, even though it wasn't a huge problem. If it had been a huge problem, all of the things that we have been doing for the last decade put us in a good position to be prepared for the next flu epidemic.

Dr. Weinstein: Thank you to our panelists and thank you for participating.