By Kelly M. Pyrek
Can aggressive isolation practices and contact precautions make a significant impact on infection rates? Should these stringent interventions be applied to all patients merely colonized with multidrug-resistant organisms (MDROs)? Those are the questions facing infection preventionists as they struggle with translating evidence into practice in the real world, where textbook scenarios arent the norm and healthcare epidemiology may not have all of the answers.
While contact precautions have been traditionally applied to patients infected with MDROs, this intervention has been expanded to include those colonized with all MDROs in many healthcare facilities. Although numerous recommendations and guidelines promote the use of contact precautions, the data dictating this practice remain resolute to some and ambiguous to others. At the Fifth Decennial International Conference on Healthcare-Acquired Infections held in Atlanta in March, experts David Pegues, MD, and Kathryn Kirkland, MD, reviewed the scientific data supporting use of contact precautions and attempted to reconcile differences in an informal debate.
Pegues, of Ronald Reagan UCLA Medical Center, and Kirkland, of Dartmouth-Hitchcock Medical Center, discussed whether "all patients colonized with multidrug resistant organisms should be placed in contact precautions." Before the onset of the debate, a select number of attendees were queried as to what they believed about this aforementioned statement and were allowed to respond using an automated polling device; 49 percent responded "definitely yes," 22 percent said "perhaps yes," 6 percent were "neutral," 9 percent said "perhaps no" and 14 percent said "definitely no." After both sides of the issue were presented, Decennial audience members voted again on the statement, "all patients colonized with multidrug resistant organisms should be placed in contact precautions." This time, 31 percent said a "definite yes" while 18 percent said a "definite no" and 4 percent remained neutral on the topic.
That practitioners were able to be persuaded to change their minds on the issue suggests an ongoing struggle with the degree to which some interventions must be implemented, and potential ramifications for patients.
"I think people are struggling with the potential benefits and the potential risks in increased isolation," says Pegues. Regarding the impact of contact precautions on patients, Pegues acknowledged that it can decrease healthcare worker/patient contact, trigger delays in care and lower patient satisfaction overall. Mitigation strategies would include better staff education and using strategies to reduce the feeling of social isolation. "I believe the potential benefit is a direct one for all patients in a healthcare facility, but maybe only an indirect one for a patient who is colonized with an MDRO; the direct benefit is that others presumably will have a decreased risk of acquisition of the MDRO and a decreased risk of developing subsequent infection. The potential risk is felt more directly by the patient who is placed in contact isolation with increased possibility of social isolation and decreased contact with caregivers. The broader perspective suggests its a valuable strategy because its going to protect a greater number of people. The individual perspective suggests there may be deleterious effects of isolation. As a public health practitioner, my perspective supports the good of the many over the good of the few; however, I do think there are ways to mitigate the potential deleterious effects of isolation on patients."
In his pro-contact precautions presentation, Pegues declared that there was a scientifically sound rationale for contact precautions to protect against the transmission of multidrug-resistant organisms (MDROs). Pegues explained that there is consistent evidence that control strategies, including contact precautions, reduce the incidence of MDROs, such as that outlined in the 2006 MDRO guideline produced by HICPAC, and that since those guidelines were published, more than 30 additional studies support this intervention. In regard to the role of gowns and gloves, Pegues says that gowns and gloves are more effective than universal gloving in the transmission prevention of MRSA and VRE. He acknowledged researcher Charles Huskins STAR-ICU study in which he determined that when looking at the incidence density of new colonization vs. infection, there is no additional benefit of active surveillance cultures and contact precautions for colonized patients. He said that there are limitations to quasi-experimental studies such as Huskins and that in nearly all studies reported that sufficient MDRO control required at least seven or eight different interventions concurrently or consecutively.
Pegues says that he has yet to be swayed from the position he took at Decennial. "I think we are all very anxious to see Charlie Huskins STAR-ICU study published in a peer-reviewed journal," Pegues says, adding, "Its been a number of years since the abstract was presented, and I think that will shed some light on this issue. And I think increasingly, as were being mandated on a state level to report HAIs and in many cases such as California, to perform active surveillance cultures, its maybe not sufficient to only isolate patients or to screen patients and then isolate them, there has to be something that comes out of that approach and whether thats based on a change in ICU bathing practices where you are using skin antiseptics like CHG or some other active approach to decolonization. I think that if you are screening, and if you are isolating patients, then if possible you should be making attempts to decolonize them. That strategy not only prevents potential transmission, but from the societal perspective -- it is going to benefit the individual because I think there is mounting evidence that once you are colonized, you increase that risk of developing infection with MRSA, VRE or bloodstream infection."
In summary of his argument at Decennial, Pegues noted the presence of unresolved issues relating to contact precautions, including when to use gowns and gloves vs. gowns or gloves alone; when to implement contact precautions in the active surveillance culture process; the duration of contact precautions and when to clear patients; universal gloving vs. contact precautions; and how to optimize contact precautions as an infection prevention and control intervention. At the very least, Pegues said, glove use and hand hygiene is the minimum of care.
Despite what is already known about these traditional interventions, experts like Pegues acknowledge that more research is a welcomed development. "There is a sense of pride that the field of healthcare epidemiology is moving forward into the realm of conducting multi-center, randomized, controlled clinical trials, that they are receiving increased emphasis as well as scrutiny, and that the overall quality of the evidence base continues to increase," Pegues says. "The minor frustration, which is not unique to this question or to healthcare epidemiology in general, is that when you conduct large, complex, randomized clinical trials with multiple authors, that there can be delays related to how the data is analyzed, how it is presented and how it is interpreted. You live with this process and you look forward to the day when the data will come out."
In her opposing argument against excessive contact precautions, Kirkland made the point that in the statement, "all patients colonized with multidrug-resistant organisms should be placed in contact precautions," the word "all" was too broad, the word "resistant" was too narrow, and that "contact precautions" was the wrong intervention. Instead, she asked Decennial attendees to consider the statement that all patients are colonized, all patients are managed in ways that reduce HAIs, and that some patients should be placed on contact precautions. As a reminder, Kirkland reiterated the indications of contact precautions as a private room, gowns and gloves worn by the healthcare worker as a barrier, and applied to one patient to protect another. Kirkland noted the evidence that contact precautions are necessary, but asserted that many are weak studies with potential biases and mixed conclusions making it easy to use them to argue for contact precautions. She said that at best, the evidence for contact precautions was weak and ambiguous. Instead, she pointed to evidence that contact precautions are not always necessary, and asked the audience to consider that contact precautions for colonized patients are not needed in some settings.
For patients not yet colonized with an MDRO, Kirkland noted that a private room can prevent transmission from shared equipment, that gloves can prevent transmission if healthcare worker apparel is contaminated, and that gloves can continue hand carriage of pathogens by healthcare workers. For colonized patients, Kirkland said that a private room cant isolate a patient from themselves; that a gown could transmit infection, and that gloves are as likely as hand to be a source of contamination and cross-infection. She added that at least 18 percent of gloved hands become contaminated during patient care, and that dirty gloves are at least as dangerous as dirty hands. "A healthcare worker with clean hands may actually be safer than one with gloved hands," she said. Regarding the impact of contact precautions on patients, Kirkland noted that the jury is still out on whether healthcare workers more easily pass by isolation rooms and that these patients receive less care than those who are not in isolation.
Following her presentation at Decennial Kirkland says she heard from people who are using the same approach in their institutions who wanted to share what they have been doing; others told her they thought what she said makes sense and gives them a rationale for making changes at their facilities.
"My approach to infection prevention has always been fairly simple and derived from common sense, and I think there are people out there who have felt there is some disconnects between what seems common sense and what the data supports," Kirkland explains, addressing the reason why some Decennial attendees changed their votes on the issue in the instant poll. "In some ways I think the guidelines may be a little more out of the box than I am and so I think the reason why people are movable on this issue is because there is a certain rationality of the argument against an organism-specific approach to infection prevention and because practitioners want to do what they feel is of most benefit to the most people. When they hear someone saying that is actually legitimate, I think that is where they move. Its hearing that someone else is willing to stand up and say this is what we do and this is why."
Kirkland also addressed the problem of putting a focus on MDROs to the exclusion of all other pathogens. She said that studies point to a smaller number of infections caused by MDROs and that if only MRSA or VRE is targeted, what happens to everything else? She said that after all, all patients are colonized with organisms that could trigger an outbreak at any time. She advised that clinicians put contact precautions in context and use them when they make sense only. A patient should be placed in a private room if there is strong likelihood that there could be widespread contamination (such as when a patient has diarrhea) or a long persistence of an organism; to use gloves only to protect the healthcare worker; and to use gowns when there is anticipated exposure to infectious material. Kirkland said it is essential to rephrase the question of what is appropriate, for whom, when and why, and acknowledged that it is a complex social and behavioral issue as well. "Its time to reconcile our differences and move toward consensus," she said, adding that to do so, the issues of determining the added value of contact precautions and the impact of reduction vs. cost increases must be addressed.
"If you really understand the approach I was taking during the debate at Decennial, in some ways its certainly not doing less; in some ways its doing more, so I think part of the appeal is that its an approach that has the potential to protect more patients, not just those who may be colonized with multidrug-resistant organisms, which is a small minority, so maybe it makes people feel empowered to do more."
Kirkland says that her facility has very low MRSA rates without having to isolate patients, and Pegues says he does not believe it is definitive evidence that this strategy will work everywhere. "In many instances I might caution against it if your rates of infection are high," Pegues says. "I think you have to base decisions about whom to isolate and how to isolate you have to base the decision on high-quality evidence and local factors."
Kirkland says that other facilities have similar approaches to Dartmouth and are seeing controlled infection rates because of it. For healthcare institutions still struggling with HAIs, Kirkland advises, "If you have a type of infection that is out of control, you need to understand what is driving it is it the environment that is not clean enough? Are healthcare workers not washing their hands enough? Is it a different way of suing antibiotics that actually allows the emergence of resistance? I think we are in an enviable place because we are in northern New England and we may have a little more control over our environment than others; some of it is the false sense that the program that you have on the books is whats responsible for the infection rates that you have its not knowing which elements of our program are actually necessary and so I think just knowing that a range of programs can result in a similar degree of control is a good starting place."
Kirkland attributes the variability in infection prevention results to the gap between policies and practices. "In my Decennial talk I made the point that people dont report on actual compliance with the policies they are examining; instead, they say, I know what the hand hygiene rates are and I know what the infection rates are, therefore I believe that in the context of high hand hygiene rates that the infections are controlled. I think when people are continuing to see high rates of infection despite aggressive policies, then the first thing I would look at is how policies are being followed by healthcare workers. Often, contact isolation isnt being followed that well and neither is hand hygiene. Do you need both? If I were to go after one of those I would go after the hand hygiene first. I think within SHEA there is the move toward more collaboration, more research across various centers, putting together a research network and I think we will be able to use that to explore some of these questions, such as why is there this variation and what drives it and what can we learn from it."
Despite the variation in infection rates across the country, cultivating a consensus on contact precautions on a national level is not something Pegues believes is needed; as he explains, "I do believe there is a need to look at the best available data, determine what the problem pathogens are at your institution, and what the most effective strategies available to control them. If you work as a hospital epidemiologist at Dartmouth and you have a low prevalence of MRSA infection and you are able to maintain that low prevalence of colonization and infection with your current control strategies that dont involve the routine use of contact precautions thats great, but I think when you have a problem you need to bring all available strategies in a thoughtful way to bear on controlling them. I am much less interested in a national consensus although I think you want to make national recommendations based on best available evidence. The compendium of strategies for the control of HAIs from SHEA and IDSA do suggest that a tiered approach is appropriate meaning there is a minimum that you should do and there are additional measures that can be brought to bear when you are not getting your desired results and you have looked very carefully at the process and the process measures."
"If there were a national consensus, what I would like to see agreement on is that there is a range of tools we have and that different ones are appropriate for different settings and that people should use data around what their problems are to help them decide which tools to choose," Kirkland says. "I think consensus on either contact precautions is necessary for all patients colonized with multidrug-resistant organisms or its unnecessary for all of them I think to get consensus you have to frame the question correctly and I dont think a yes-or-no question is going to ever result in consensus."
Absent such a national consensus, a greater grasp on translating evidence into practice is desirable, especially on the part of infection preventionists who endeavor to help healthcare professionals at the bedside use interpret and use data.
"It is my perspective and that of HICPAC that the category 1 recommendations are category 1 because they are based on strong epidemiologic or direct clinical evidence," Pegues says. "But within those recommendations there have to be priorities made. I think we remain in the realm of finding out what the best strategies are -- doing randomized clinical trials or observational studies using quasi-experimental design -- we are moving into the area of implementation science, based on what these recommendations are, what is the best way to achieve the desired results and that is encouraging. I dont think there is necessarily confusion; the recommendations are what they are as far as the issues, but in a real-world sense, there is still the question of how best to implement these recommendations, to measure success toward meeting them and achieving the desirable goals. Its exciting that healthcare epidemiology and infection prevention is moving toward implementation science.
It may feel as though some unresolved issues -- such as when to remove patients from contact precautions -- persist, and Pegues says the bottom line is not necessarily to find the optimal duration for contact precautions but that the ultimate goal is to "decrease individuals risks of acquiring colonization or infection with MDROS associated with medical care." Pegues adds, "I am less concerned about unresolved issues and more concerned about the outcome."
Now that SHEA has announced the launch of a new research collaborative, perhaps translating evidence into practice could be facilitated more easily. As Kirkland notes, "The consortium will most likely make an effort to make evidence more practical. I think to the extent that it tries to tier things a little there are basic interventions and things to consider if your problem isnt under control is a good start in acknowledging that the context in which you are working makes a difference in what you do. I would like to see a little more movement in that direction so that we can actually try to understand what elements in that context are important in determining whether you need to go to the higher tier or not. There are so many evidence-based or pseudo evidence-based practices we can choose from; certainly not all of them are necessary all of the time and so what I would love to see as the next area of exploration is to decide which set of these practices makes the most sense for the particular situation allow people to determine what makes sense rather than saying, These are the 500 best practices and you must implement all of them every time. That cant possibly be necessary."