Hand Hygiene Interventions: A Network Meta-Analysis

Summer is in full blaze (especially for those in Rome, France and the western US), so we don't have much time for long posts. However, I had to point you to an excellent study in the BMJ (open access) by Luangasanatip et al. that utilized a systematic review and network meta-analysis to determine the comparative effectiveness of the WHO 2005 hand hygiene campaign and other interventions. The WHO-5 Campaign (not to be confused with the WHO 5 Moments) recommended a multimodal strategy consisting of five components: system change, training and education, observation and feedback, reminders in the hospital and a hospital safety climate.

The authors completed a systematic review of interventions from 2009-2014 and used prior reviews to identify other studies. A strength of the analysis was that they looked beyond randomized trials and included high quality quasi-experimental studies including non-randomised trials, controlled before-after trials, and interrupted time series studies. They then completed a network meta-analysis which suggested that the WHO-2005 campaign was effective and compliance could be improved if other interventions were added including goal setting, reward incentives and accountability.

For those interested in reading more about network meta-analysis, I suggest you read John Cornell's editorial and the PRISMA Extension Statement in this past June's Annals. Briefly, it allows direct and indirect comparisons of interventions. For example, if two interventions are not directly compared they can still be compared if they were both directly compared to a third intervention (see Figure 1 above - Treatment D vs Treatment B or C through their direct comparison to Treatment A). Additionally if there is a closed-loop of studied interventions, additional information can be gained from indirect comparisons even if direct comparisons also exist. For example, in Figure 1 above, we can learn about Treatment A vs Treatment B from their direct comparison but also indirectly through Treatment C.

I encourage you to read the full study and the editorial by Matthew Muller. Very nice to see that the BMJ published this important study. And for those in the southern hemisphere, enjoy your cool weather...these summers seem to be getting worse and worse.

Clothing and corporate culture

Anyone who is a regular reader of our blog knows that healthcare worker clothing is a favorite topic, from the viewpoints of both infection control and "professionalism." And as we've recently blogged, colleagues at the University of Michigan are trying to ramp up professional attire, calling for doctors to put their white coats back on. But they've been outdone by Summa Health System in Ohio. Summa has now mandated that all healthcare workers at their hospital must wear underwear. That's right, no more going commando at Summa! [I did not make this up--see here]. Now as a pragmatist I have to wonder who is in charge of inspection and enforcement of that policy and exactly how they will inspect and enforce.

All of this reminds me of that classic SNL skit where Will Ferrell does indeed wear underwear to work.

I guess it's all about corporate culture. One company well known to many healthcare personnel is the behemoth EMR vendor, Epic. Its corporate culture makes it a place that is well known for employee engagement and friendliness. Having visited the Epic campus recently, I can attest to the palpable enthusiasm of its workers. Epic's dress code? When there are visitors, you must wear clothes. No mention of underwear, though.

Today's lesson: Take your zoo off and put your smile on

I ran across two great quotes regarding the white coat today:

• In a piece in KevinMD.com published today, Shivam Joshi, an internal medicine resident, writes about doctors wearing their white coats in grocery stores. He describes the white coat as "20 square feet of a microbiological zoo."
  
  
• Over at BMJ, Edmond Fernandes argues in an essay that physicians in India should stop wearing white coats. Here is the money quote: [White coats] are mere symbolism and wearing them does not itself confer status or professionalism. Dressing presentably and sporting a smile are more important than white coats.

Killing antimicrobial stewardship (part 2)

A few days ago, I posted a piece regarding the 21st Century Cures Act, which would provide incentives to hospitals to use new antibiotics. This is terrible policy, in my opinion. Through comments on Twitter and this blog, as well as conversations with colleagues, there was concern raised that my cited source was incorrect. While the New York Times is not infallible, it's pretty damn good. In all honesty, I read the text of the bill myself before posting but found it unintelligible (maybe I was too tired at the time).

Anyway, a highly reputable group, the Center for Disease Dynamics, Economics and Policy (CDDEP), has bailed me out. They cite the same sentence from the New York Times editorial that I did in my previous post, and add, "We at CDDEP couldn't agree more."  Even better, they argue that stewardship not development of new drugs is the ultimate solution to the antibiotic resistance crisis. I highly recommend their post, which you can find here.

Is antimicrobial stewardship about to be killed?

While still half asleep this morning, reading the New York Times, I saw something so disturbing that I almost choked on my Pop Tart. In an editorial about a bill that has passed the US House of Representatives, I read the following:

[The 21st Century Cures Act] would expedite the use of new antibiotics by providing financial incentives to hospitals to use them — benefiting manufacturers but also driving up costs and encouraging overuse, potentially breeding resistant superbugs.

It's hard to imagine that in the face of the alarming issues we are having with antibiotic resistance, and all the efforts hospitals across the country are expending to promote antimicrobial stewardship, that our government is entertaining paying hospitals to increase the use of new antibiotics. If this bill becomes law with this provision, antimicrobial stewardship is dead. What could possibly be next?

Bacteria and Viruses to Humans -> I Am Invisible

Last week, in typical Controversies Blog fashion, we had a bit of a dust up discussing white coats and presenteeism. It occurred to me that in each instance, the primary barrier to infection prevention remains our inability to recognize that our adversaries are invisible. How else can we explain doctors insisting that washing white coats every three days is adequate and that they've never personally transmitted pathogens while caring for patients? So for those of us that still think hand hygiene compliance of 40% is satisfactory, contact precautions are bogus and that coming to work sick is cool, I have a three word reminder: pathogens are invisible. ciao

Are there too many ID specialists?

As ID fellowship interview season begins, there have been some interesting posts on the program directors’ list serve. After last year’s historically awful match, program directors are looking for answers, and finding none. A longtime program director opined that “there are too many ID programs”, referring back to the famous words of Dr. Robert Petersdorf that we’ll eventually end up “culturing one another”. This individual argued that ID programs should drop their research emphasis and instead train fellows in the skill sets they’ll need in community practice (infection prevention, stewardship, clinical microbiology, basic business principles, etc.). Furthermore, the argument went, the reason IDSA has been slow to respond is that the society leadership is over-represented by academic ID physicians. 

Now, it is objectively true that we have too many ID training programs when compared with the existing demand for training….but I frankly hadn’t considered the argument that we simply have too many training programs compared with actual future need for ID physician expertise. So I went back and pulled the piece in which Petersdorf made his argument that we are training too many ID docs. Interestingly, his argument about how ID programs should adapt to future needs was the exact opposite of that being proposed by the program director mentioned above. Petersdorf argued that:

“...the existence of specialists in most community hospitals will lead to fewer referrals to the teaching centers and the resulting lack of patients will lead training programs to atrophy. Infectious Disease is destined to function best as an academic specialty whose trainees should pursue careers primarily as investigators. The number of clinicians leaving training should be reduced and not further glut the marketplace; they should be based in academic divisions and devote their clinical time and effort to the care of complex referrals and to indigent patients.” 

So which is it? Do we have too many training programs (and thus are counter-productively trying to gin up interest in ID when we don’t really need more ID docs)? Or are we destined to face a critical shortage of ID specialists in the near future?

In my view, the answer depends upon answers to other questions that the current trainee has very little control over, including: (1) will the funding climate improve for physician scientists?, (2) will health care reimbursement in the U.S. eventually move fully from RVU-based to value-based?, (3) if the answer to question 2 is yes, how will that health care be delivered?, and (4) will ID physicians continue to play a major role in individual hospital quality-safety programs (e.g. infection prevention, antimicrobial stewardship), or will their roles be supplanted by hospitalists and other non-ID trained quality-safety experts?

Expanding on questions 2 and 3 above, I’ve often heard it said that value-based purchasing and the move away from RVU-driven reimbursement would be a boon to ID docs. I’m not so sure about that. Many areas of our rural state have no local ID expertise—as a result, our faculty field several calls per day from providers seeking (free) ID expertise, are now performing “eConsults”, and plans are for us to greatly expand our telehealth programs. These programs are designed not, as Petersdorf envisioned, to increase “referrals to teaching hospitals”, but instead are meant to reduce the need for these expensive referrals.

What is the least expensive way to provide ID expertise to a large population? Sadly, I envision a future “ID Command Center”: an advanced telehealth unit that allows one ID doc to provide consultative support to entire healthcare networks, providing instant access to all the required information….except for that which can be gained by actually touching the patient. Sorry, Abraham.

What do you all think?

Photo credit: Mike Staugaitis

We need to rethink professionalism

A new study in JAMA Pediatrics (free full text here) should make hospital epidemiologists and infection preventionists cringe. All physicians and advanced practice providers at Children's Hospital of Philadelphia were sent a survey on presenteeism. Of the 929 providers surveyed, 58% responded.

The big findings were as follows:

• 95% felt that working while sick puts patients at risk
• 16% would come to work with fever 
• 30% would come to work with diarrhea
• 5% would come to work with vomiting
• 56% would come to work with acute onset of respiratory tract symptoms
• Asked several ways, >90% stated they would come to work while sick out of a sense of professional obligation

What I find most interesting about this paper is not how often sick healthcare workers come to work (though it's a big problem), but why they do so. And once again, professionalism rears its ugly head. Professionalism revolves around expectations and norms set by the profession. It seems to me that if we viewed this more through the lens of humanism rather than professionalism, we'd be better off. From a humanistic standpoint, which holds a universal rather than parochial view, all would agree that individuals who are ill with potentially contagious diseases should not come to work (doesn't matter whether you're a doctor, a teacher or a plumber). Similarly, in the white coat debate, professionalism drives the argument that physicians should wear a white coat. Humanism would dictate that attire should be practical, comfortable, safe, and personally desirable. Several years ago, Judah Goldberg wrote a great essay in Academic Medicine that I often quote and recently re-read on the conflict between humanism and professionalism. It really crystallized for me the differences between the two philosophies, which are often in conflict, and once the differences are made clear, it's easy to see the corrupting influence of professionalism. It's worth a read.

The white coat debate continues on and on and on and on...

I couldn’t resist blogging in response to Eli’s recent post regarding doctors’ attire. I guess this debate continues on with some new warriors from the University of Michigan. They plan to do a bigger, worldwide study of patient preference for physician attire. We really do need another meaningless study on this topic, don’t you think? While they’re at it, I think they should ask patients their favorite color, favorite flower, and favorite season of the year, because it would have the same impact.

I have blogged on this topic many times (see here, here, here, here, and here) and I won’t rehash all the arguments in this post. But these are the most important ones, I believe:

1. I think we need to act consistently about the role of clothing in infection control. That is, if you believe that contaminated clothing plays no role in transmission of infection, then be consistent and eliminate contact precautions. If you do believe that clothing may transmit pathogens, ditch the white coat and employ bare below the elbows or contact precautions, or some combination of the two. Or if you believe that the white coat magically resists contamination by pathogens, disclose that as well and those of us who believe in the germ theory and other scientific concepts like global warming can move on.
2. If you think that we need more data about patient preference then put the issue of physician attire into context in your survey or test your hypothesis with a clinical trial. Both types of studies have been done. And the results are clear: when placed into context, patients find physician attire to not be very important (not really surprising—most patients would rather have a kind physician who listens well and wears scrubs than a mannequin in a white coat) and when formally tested in clinical trials, attire had no impact on patient satisfaction with their care (see here and here).
3. The white coat is all about the doctor, more specifically the doctor’s ego. It’s truly about professionalism in the most negative sense of that concept (physicians judging other physicians to “protect the profession”).

In my current job as Chief Quality Officer at an academic medical center, I am able to separate my time fairly cleanly into clinical time and administrative time. When I am seeing patients, I wear scrubs, have done so for the past 6 years, and have never had a complaint. Interestingly, when my CEO sees me in scrubs, he always points to the scrubs and comments, “that’s a good look!” I also don’t consistently introduce myself as “Doctor,” particularly in situations where I am likely to follow a patient over a long period of time, and in cases where the patient’s status is tenuous and they need to be able to reach me quickly, I give them my cell phone number. Call my crazy, but this approach to patient care works well for me. When I’m working in my administrative role, I frequently wear a tie, though not always, rarely wear a sports jacket, and almost never a suit. Most physicians have enough common sense to dress appropriately for work, so very few need to be told how to dress. And I respect the fact that each of us has our own style. How you dress is a personal decision, a reflection of who you are. By the same token, there are some patients who prefer their doctor to dress more casually. There is no one size fits all.

So to our colleagues in Michigan, a challenge: in your new survey ask patients the following question: which is more important to you—that your physician be dressed in a white coat or that your physician gives you his/her cell phone number? I can’t wait to see the results.

The limitations of patient-centered infection control

"You can't just ask customers what they want and then try to give that to them. By the time you get it built, they'll want something new." - Steve Jobs

I woke up to Vineet Chopra and Sanjay Saint's editorial advocating for a patient-centered approach to selecting clinician attire. In the editorial they point out that in the 21 of 30 papers they reviewed "patients had strong preferences about what physicians wore. And it looks like patients more often prefer for their doctors to wear formal clothing and white lab coats than not." They also reference the SHEA guidance document on healthcare attire and state there is little evidence "that germs on male doctors’ neckties, long sleeves, or white coats actually spread infections in a nonsurgical setting. So bans on such garments, such as those in place in some countries, may go too far."

So how can we decide what the safest attire is for our patients? I would recommend epidemiological studies that track bacteria spreading in hospitals and attempt to determine if those same strains are contaminating clinician attire using whole genome sequencing.  In fact, CDC released a SHEPheRD Task Order 2015-006 last month that seeks to do just that and more. I very much look forward to the results.

Drs. Chopra and Saint recommend a different patient-centered approach as they "plan to survey thousands of patients from the US, Italy, Switzerland and Japan" and "will specifically assess how factors such as age or how often a person interacts with the health system shape patient opinions." Is this a rigorous or unbiased method? Will the surveyed patients know that a quarter of white coats are coated with invisible S. aureus or MRSA and that they are rarely washed? Will they understand that a lack of data supporting transmission from white coat or long sleeve or neck tie is because no one has ever funded such studies? Will they understand that the circumstantial evidence supporting bare below the elbows is as strong as that supporting a clean environment in hospitals? Will their next editorial say that bleach is stinky and irritating to patients, so we shouldn't use it in hospitals unless supported by a patient-centered survey?

Thus, there are several potential limitations to patient-centered infection control and the planned physician attire survey, which I hope they will consider before collecting data and selling the findings. I'm all for patient-centered healthcare where applicable and data are fairly presented to patients and I support evidence-based medicine when we fairly rate the science based on what studies have been funded or will ever be funded. I'm not sure any patients "feelings" about the dirty white coat are worth the increased risk of MRSA or Acinetobacter infection, but we can disagree. All I would ask is that if we do patient-centered research, it's with properly informed patients.

***
Oh, and if we're going to require evidence before making physician attire recommendations, how can the authors write that "scrubs do not belong outside the hospital environment. Especially not in the grocery store." Where is the study that scrubs spread bacteria outside of hospitals and that there is any risk to population health in grocery stores? Clearly the bacteria on cantaloupes are riskier than those on scrubs! And surely the immobile ICU patient with central venous and urinary catheters is at greater infection risk when you wipe your MRSA-contaminated white coat on their catheter than when you brush up against a healthy grocery store patron with your scrubs? Scrubs are at least washed daily - white coat laundering occurs somewhere between every two weeks and... never.

*****
Additional thought, I am not aware that the US, Italy, Switzerland and Japan have mandatory bare-below elbows policies. It might be important to add Scotland or England to the survey since they have had BBE policies for some time and patients may have adjusted their preferences for physician attire.