Another use for the checklist

A press report from the Society of Critical Care Medicine Meeting outlines a study with a simple, novel intervention to improve hand hygiene compliance. In this study, investigators in a surgical/trauma ICU added a question to their daily care checklist: Has anyone seen anyone else touch the patient without washing their hands in the past 24 hours? If the answer is yes, the name of the offender is recorded. The result of this simple intervention was improved hand hygiene compliance from 69% to 89%. Though there isn't much detail about the study given in the report, one could envision how this would integrate front-line providers into a continuous vigilance that could be quite powerful. One downside is that this would likely not work well outside the critical care setting.

Breaking down the CLABSI bundle

There is an interesting CLABSI prevention study out this week in PLoSONE (full disclosure: tireless co-blogger and pal Eli is one of the authors). In order to determine how well the CLABSI bundle was being implemented in U.S. ICUs, and which individual elements (or subsets of elements) were most strongly associated with CLABSI reductions, the authors surveyed NHSN hospital practices. They used quarterly ICU-specific CLABSI rates as their outcome measure.

The bottom line: the bundle was associated with lower CLABSI rates only for units that monitored and reported high rates of compliance with at least one element of the bundle out of three (maximum sterile barrier precautions, optimal site selection, and daily assessment of need).

Why would meticulous adherence to any one of these three bundle elements be significantly associated with CLABSI reduction, while meticulous adherence to all the elements was not? I believe it was a simple power issue: too few ICUs (only 38%) had high rates of adherence to the whole bundle. Lack of power could also explain why no single element was statistically-significantly associated with CLABSI reduction.

This study is a good first step toward “breaking down the bundle”, to determine which elements are most important for infection prevention, what compliance measurements are most useful, and (eventually) what components should be added, or subtracted, from existing bundles.

Another checklist victory

There's a new paper in BMJ on the use of checklists. In this study 8 care bundles, each with an associated checklist (available on the BMJ website), were introduced in the hospitals of the North West London Hospitals Trust. Five of the bundles were related to prevention or management of infections (central line insertion, ventilator associated pneumonia, MRSA, diarrhea and vomiting, and surgical site infections). Changes in mortality were monitored for diagnoses that would be expected to be impacted by the bundles as well as overall mortality. Significant reductions in mortality were noted beginning one month after implementation of the checklists. It is important to point out that this is a quasi-experimental study and confounders may be at play. Nonetheless, it's another piece of evidence that supports the checklist concept. After having now read both Peter Pronovost's and Atul Gawande's books on checklists, I am convinced that these simple tools can have huge impacts by their ability to drive high levels of compliance with practices that we know reduce risk. Ironically, the implementation of a simple and effective checklist turns out to be enormously complex given the culture of medical care which has traditionally bowed to doctor's autonomy (i.e., their egos). I was talking to one of my favorite surgeons last week about these issues, and he summed it up well: "The young surgeons get it. As for the old ones, I think we'll have to wait for them to die off."

Safe Patients, Smart Hospitals

I had some beach time last week so I read Peter Pronovost's new book, Safe Patient, Smart Hospitals. It's the story of his journey in patient safety, which starts with his father's death, likely hastened by a medical error. Parts of the story are probably familiar to those who work in infection prevention, but I think it's worth reading.

One of the major points he makes is that the checklist, while important, can really only work when the hospital unit embraces a culture of safety. An aspect of the book that I particularly liked is his criticism of some of the work in quality improvement because measurements lack validity. Like Pronovost, I've been accused of trying to do research by QI folks, when all I was asking was to measure a process or outcome accurately and precisely. He also points out how often ego gets in the way of doing the right thing. Over and over, I kept wondering why change is so difficult in  hospitals even when the data for a new intervention are compelling. And he reinforced my belief that infection prevention, like many other aspects of patient safety, is all about high levels of compliance with simple practices.

Unfortunately, I suspect that Dr. Pronovost is preaching to the choir. Those who might benefit most from his words are least likely to turn the pages of this book.