Tuesday, March 31, 2009

Wireless Monitoring, Real-Time Feedback?

Feedback of hand hygiene rates is a key part of any hand hygiene campaign—unfortunately, such feedback is often delayed, and delivered at infrequent intervals. Feedback delivered in real time is most effective—the best example being a gentle reminder from a coworker at the bedside.

A couple weeks ago, my colleague Phil Polgreen presented, at the SHEA Annual Meeting, the results of a pilot study of a new wireless hand hygiene monitoring system he developed in collaboration with a team of computer scientists here. The potential for real-time wireless monitoring got me thinking about ways to provide real-time feedback as well. Such a system could be programmed to feed simple aggregate data wirelessly to a large wall display—say, just the rate of hand hygiene dispenser use upon room entry or exit. The display could be patterned after the traffic signs placed by public safety officers on stretches of road prone to speeding:

I’m sure there would be plenty of glitches, ways to game the system, etc., but I wonder if having this information “in your face” all the time, continuously updating, would have an impact?

Making Contact Precautions More Friendly

In a recent posting, Dan Diekema pointed out some of the problems with contact precautions, one of which is a significant reduction of physician and nurse visits to isolated patients.

When our ICUs recently moved to a new building, one of our infection control nurses came up with a great idea. She noticed that in each patient room in the building, the floor color of the first two feet from the room entrance is different (blue) than the remainder of the floor (tan) as shown in the photo.

She proposed that for patients in contact precautions, gowns and gloves need not be worn when standing in the blue area and items in this area (e.g., the over bed table used for charting) are considered clean, but gowns and gloves must be worn when stepping onto the tan floor in the room. This has had several positive effects: nurses can use the overbed table on the blue floor for documentation without donning gowns and gloves, it allows for greater observation of the patient and for staff to speak to the patient more easily, and it is environmentally friendly by reducing the use of consumable products. Given the success of this change, we are now making similar changes in older parts of the hospital as they are renovated. This is a nice example of a simple solution that facilitates good care, reduces costs, and is green.

Monday, March 30, 2009

More on Industry Conflict of Interest

Just one week after Kathy Kirkland’s enlightening and thoughtful talk on industry conflict of interest at the SHEA Meeting, JAMA has published proposed recommendations for controlling conflict of interest with industry for professional medical associations. The authors note that the recommendations are rigorous and will likely be controversial. Nonetheless, these recommendations should be embraced. While physicians are generally aware of the role of industry in the arena of clinical care, it seems that in infection control industry’s role has gone largely unnoticed. Little has been written in the main stream media about this, with Andrew Pollack’s article in the New York Times last year a notable exception. Ted Eickhoff and I also wrote about the role of industry in the MRSA active surveillance debate in CID, which was subsequently criticized by APIC in a letter to the editor. Hopefully more attention will be drawn to this issue, which will raise awareness, a first step in reducing industry’s influence.

New CDC Guidance for Carbapenem-Resistant Enterobacteriaceae

The CDC just issued new guidance for labs and infection control programs to help detect and control carbapenem-resistant Enterobacteriaceae (CRE).

You can read this guidance for yourself, but the quick and dirty summary is: (1) micro labs should follow the latest CLSI guidelines to detect carbapenemases, which means doing a modified Hodge test on all isolates with elevated carbapenem MICs; (2) all patients infected or colonized with CRE should be placed in contact precautions; and (3) all hospitals should review up to a year of lab data for unrecognized CRE, and perform point prevalence surveys in high risk units if CRE if found. When CRE is detected, epidemiologically-linked patients should be screened for CRE—with continued screening until there is no evidence for cross-transmission. Additional measures are consistent with those already recommended in the CDC’s guidelines for control of multidrug-resistant organisms.

I’m of two minds about this new CDC guidance….on the one hand, I welcome any clear recommendations for control of multiply drug resistant gram negative rods (MDR-GNRs). On the other hand, CRE is only one of many MDR-GNRs, and the screening method currently recommended by the CDC is CRE-specific and is of uncertain sensitivity in clinical practice—many hospitals are also battling garden-variety ESBLs, AmpC-producers, MDR-Acinetobacter and Pseudomonas, and others. There is already evidence that screening for MDR-Acinetobacter (even with multi-site sampling), is poorly sensitive. So a negative screen can be falsely reassuring—it may not exclude CRE carriage, it doesn’t exclude carriage of other MDROs, and it represents only one point in time (and unfortunately, patients don’t become MDRO colonized at 3 or 7 day intervals for our convenience). Remember when we used to think that swabbing a nostril and plating to solid media would detect almost all MRSA carriers?

As resistant organisms accumulate, with increasingly complex patterns of resistance and colonization, we will need to revisit our assumption that actively seeking carriers of the “bad bug du jour” (and placing those carriers in contact precautions) is the right solution. Are we prepared to perform multi-site sampling tests (that are of questionable sensitivity) to assess for asymptomatic carriage of MRSA, VRE, ESBLs, CRE, MDR-Acinetobacter/ Pseudomonas, C. difficile, and on and on? It might be better to take our cue from the standard precautions philosophy, and assume that all patients harbor pathogens that can be transmitted in our hospitals. Take a uniform approach to all patients, including standard precautions (including better hand hygiene!), bundled approaches to prevention of device-associated infections, liberal use of chlorhexidine (bathing, dressings, oral care), and selective use of barriers as outlined by Kathy Kirkland—and reserve active surveillance for outbreak settings.

Addendum: The last paragraph of my post above isn’t meant to imply that I disagree with this new CDC guidance….this important guidance was issued in response to what is really a nationwide outbreak (rapid emergence and spread) of a particularly bad bug. The concept of using active surveillance selectively in an outbreak setting is one with which I agree—identifying as many carriers as possible, assessing the “colonization pressure” in your population, all of this can be very helpful, even essential, during outbreaks (as long as one recognizes the limitations of the screening tests).

My point was that active surveillance as an ongoing strategy, in the endemic setting, and for the myriad of resistant organisms now present in our hospitals, is not sustainable. The other point is that active surveillance test results are too often falsely reassuring—we imagine we can identify the reservoir when in reality we cannot. We just don’t know enough yet about patterns of colonization and test performance. Finally, identifying carriers is of little use if healthcare workers aren’t adhering to good infection control practice. I can think of few more worthless practices then chasing down every MRSA carrier in a hospital where fewer than half the healthcare workers wash their hands or properly practice contact precautions.

Battling Conflict of Interest with Industry

The American Psychiatric Association Board of Trustees has voted to phase out industry-supported symposia at its annual meetings. SHEA and APIC should follow their lead.

Sunday, March 29, 2009

Hand Hygiene: Yes We Can!

As I have been rounding all over the hospital this last week on the inpatient infectious diseases consultation service, I have noticed healthcare workers reflexively hitting the alcohol foam dispensers on room entry to clean their hands. It appears that many don’t even think about it anymore; it’s just become part of what we do, much like fastening your seatbelt when getting into your car. In 2004, when we first started our hand hygiene campaign, our compliance rate was about 45%. In 2008, our hospital-wide compliance rate was 87% (over 32,000 hand hygiene opportunities observed). Details of our hand hygiene observation program were presented last week at the SHEA Meeting by Mike Stevens. Could our compliance rate be falsely elevated by the Hawthorne effect? Sure, but our 2003 rate was likely similarly affected. However, since our observers are often directed to areas of the hospital where compliance is known to be low or clusters of infections have occurred, some underestimation of the overall rate may also be occurring. Most healthcare epidemiologists believe that hand hygiene is the most important measure to prevent healthcare associated infections and I think that a strong hand hygiene program has to be the cornerstone of an effective infection prevention program. I’m a believer--we have seen a nearly 75% reduction in HAIs over the time our hand hygiene compliance has doubled, though we have implemented several other interventions in that time period that were likely also effective. There are still some who say that hand hygiene compliance cannot be driven to high levels. To that I say: Yes, we can. Yes, we have. But we still have more work to do.

Saturday, March 28, 2009

Would you like some chlorhexidine with that?

More evidence is out this month regarding the benefits of chlorhexidine in infection prevention.  This multicenter, randomized trial, published in JAMA, demonstrated that chlorhexidine-gluconate impregnated sponge dressings could significantly reduce the rate of catheter-related bloodstream infections.  This comes as other evidence accumulates that chlorhexidine bathing of ICU patients also might reduce the rate of bloodstream infection (so-called "source control"). 
A couple things are notable about this new JAMA study, however:

First, a significant reduction in BSI was found despite the fact that the ICUs involved already had very low infection rates (less than 2 per 1K cath days). Conventional wisdom has it that hospitals should first implement the standard "bundle" for BSI prevention, which includes a checklist to ensure good practice for line insertion and care.  Only those hospitals with continued high rates (defined by many as more than 2 or 3 infections per 1K cath days) should then consider the addition of more "technological" approaches (such as chlorhexidine dressings or antimicrobial impregnated catheters).  This study strongly supports the addition of chlorhexidine dressings to the prevention bundle, from the beginning.

Second, and more importantly, this is a great example of the kind of study we should be doing in infection prevention!  Much of current practice is based upon poorly designed, single center, "before-after" studies, due in large part to a lack of funding for well-designed randomized controlled trials (which are both difficult and expensive!).  This study was funded not by industry (though Ethicon did provide the dressings for free), but by the French Ministry of Health; it now represents an important addition to our knowledge, and has helped to answer a question that the most recent CDC guidelines rightly concluded was an "unresolved issue".  Bravo!

The accompanying editorial, by Eli Perencevich and Didier Pittet, is also well worth reading. 

Magical Thinking

About 10 years ago my hospital decided it needed to develop a drug testing policy for its residents. The goal was to prevent young doctors from working while impaired. As a seasoned residency director, I couldn’t make much sense of this. In my estimation, impairment at work was common among residents, but rarely, if ever, from drug use. Rather, the effects of acute and chronic sleep deprivation were widespread. Yet when I suggested we develop a policy to limit work hours rather than do drug testing, I was treated as if I were insane. It took another decade and reports and regulations from the Accreditation Council on Graduate Medical Education (ACGME) and the Institute of Medicine (IOM) to convince doctors what the average adult human being has known since the beginning of time: sleep deprivation is not healthy and it affects work performance.

In 2007 when the UK National Health Service announced its ban on white coats and neckties and mandated a “bare below the elbows” approach, I thought about this for awhile and came to realize how perfectly rational was this mandate. White coats are typically not laundered daily and come into contact with many patients over the course of time they are worn. Numerous studies have documented that pathogens, including MRSA, can be cultured from them. While most healthcare workers have come to appreciate the need for hand hygiene, they think nothing of donning their contaminated coats daily and carrying bad bugs from one hospital room to another. Moreover, when confronted, many physicians, including some experts in healthcare epidemiology, argue that no one has shown that lab coats transmit infections. While that is technically true, it’s another example of magical thinking. Many physicians have a great deal of difficulty thinking rationally about their white coats. The need to wear it is deeply inculcated, and for some I think it’s inextricably tied to their egos. Despite being among the most educated persons on earth, physicians often hold deeply rooted biases that trump rationality.

At the Society for Healthcare Epidemiology Meeting last week in San Diego, my group presented two papers that should help doctors rid themselves of their beloved white coats. Dan Markley presented results of a survey of doctors that showed that roughly one-third wash their white coats weekly, another third every other week, and the remaining third wash their coats monthly or even less often. Dawn Butler presented her work showing that in the laboratory organisms inoculated onto white coats can indeed be transferred from the coat to skin with little effort.

Over a year ago, I adopted the “bare below the elbows” approach when I work in the inpatient setting. It has made me cognizant of several things: how often the skin of my forearms touches patients in the course of care, how visibly dirty many lab coats are, and how much easier it is for me to wash my hands without worrying about getting the cuffs of my shirt or coat wet.

At this point, there’s enough evidence to conclude that the potential for white coats to transmit infection is biologically plausible. And while we rarely change practice on the basis of biologic plausibility alone in medicine, a strong argument can be made to do so when the intervention poses no harm, has the potential to improve care, and costs little. Therefore, the infection control committee at my hospital has recently recommended that all healthcare workers follow “bare below the elbows” in the inpatient setting.

I’ll discuss the white coat and professionalism issues in a future posting.

Friday, March 27, 2009

Why I hate Contact Precautions, vol. 1.

Because they hurt patients. And yes, I know that using barriers can be essential to controlling outbreaks of bugs that are spread by direct contact.....but somehow the use of contact precautions has become embedded in our standard practice for managing any patient found to carry a resistant organism, whether or not they have draining wounds, uncontrolled secretions, etc.--even in non-outbreak settings.

But the evidence to support use of contact precautions outside of outbreak settings is weak. Not only that, but the implementation of contact precautions in most hospitals is even worse than that of hand hygiene.

Why am posting this? Because Kathy Kirkland, hospital epidemiologist at Dartmouth, has a great piece in the March issue of CID that proposes a new approach to the use of contact precautions. I suggest that you read it.

Dr. Jernigan is making sense! And is positively deviant....

From a NY Times blogpost by Kevin Sack, regarding the recently presented (at the SHEA 19th Annual Scientific Meeting) CDC study about using a "positive deviance" approach to MRSA prevention:

“I think this shows that hospital-wide active surveillance is not necessary to show a big decrease in MRSA....It’s not the active surveillance itself that makes the difference. It’s what you do with the information that makes it important.”