Thursday, April 21, 2016

Here’s a problem with that new 6-step hand hygiene study


A 6-step alcohol-based hand hygiene technique is significantly superior to the standard 3-step technique in reducing bacteria colony counts.

So says a randomized trial with 78 nurse and 42 doctor participants recently published online in the journal Infection Control and Hospital Epidemiology. The full text of the study is available here.

At 42.5 seconds, the 6-step process took significantly longer than the 35 second 3-step.

Multiple media outlets, including the New York Times, published stories about this study.

The study was done properly. But after reading it, I had an issue.

I conducted an informal, unscientific Twitter poll of emergency department nurses and doctors asking how many times per hour they used hand sanitizer or washed their hands at a sink. Of the 95 respondents, 43% performed hand hygiene 10 or more times per hour.


For the sake of simplicity, assume everyone switches to the 6-step technique. Factor in walking to a dispenser or sink and going on to the next task, at least 15 or more seconds will have transpired. Let's say that's about 1 minute total time for each hand hygiene event.

If a typical ED staff member performs hand hygiene 10 times per hour, that individual will spend 80 minutes on hand hygiene during a typical 8-hour shift.

The 19% of ED personnel who clean their hands 15 times per 8 hour shift will spend 120 minutes (2 hours) on hand hygiene.

In 2016, nearly all patient encounters involve wearing gloves. Is it practical for hospital workers to spend 17-25% of their shifts cleaning their hands? How will ever they find the time to check all the boxes in the electronic medical record?

In case you aren’t familiar with the 6-step technique, here’s how it’s done:
Do you think it's a bit complicated? So did the subjects of the experiment. Despite receiving personal instruction on the 6 steps and knowing they were being observed, 35% were not compliant. I would expect an even higher non-compliance rate in real life.

8 comments:

Anonymous said...

Now imagine doing this on rounds, with 40 patients to be seen, and a chief, pgy 3, and 2 interns who have to line up at the sink to do this.

Skeptical Scalpel said...

Excellent point. What about the med students? I picked the ED because I wanted a quick response.

Levi said...



Washing hands is very important. No debate on that.

Does this method have data proven clinical support for less infection transmission?

Sure it lowers colony counts but that doesn't necessarily translate to safer patient care.

Always love your posts.

NeuroTrumpet said...

Another step toward make each patient a "bubble-boy."

Skeptical Scalpel said...

Levi, I agree washing hands is important and that colony counts do not necessarily result in infection. I guess it's a lot easier to do a study like the one I wrote about than to try to prospectively follow physicians who wash their hands by different methods to see how their patient's turn out.

Neuro, I hadn't thought of that but you may be right. I don't think it's possible to eradicate all bacteria from the environment or from the caregivers.

Unknown said...

When I was in med school I remember doing a study about the use of the WHO-recomended hand washing technique in my university hospital. One of the first questions while researching this topic was if there was a specific cut-off point of CFU obtained from a hand that would tell us if it was considered clinically contaminated. This proved to be difficult.

Hand washing definitely decreases risk of infection transmission, but how much time of hand washing is truly necesary? How many CFU should we aim for? Are the clinical benefits of a 40 second hand wash really that much bigger than a 30 second hand wash?

Unknown said...
This comment has been removed by a blog administrator.
Skeptical Scalpel said...

Andres, your second paragraph sums it up nicely. As far as I know, your questions have not been answered.

Post a Comment

Note: Only a member of this blog may post a comment.