We’ve known since Dr. Semmelweiss began washing his hands between patients that good hand hygiene slows down or prevents the spread of infection. From the days I was a nurse’s aide in the 1960s I was taught to wash my hands before and after I touch a patient. I suppose I miss a few times, but I do my best to remember always to do that. Broadly speaking, though, how well do doctors and nurses do with adhering to those rules? Not so well, I’m sorry to say — usually only 60-70% of the time. It’s a bit hard to measure because direct observation to count the times in itself can serve as a reminder, goosing up the number. Then, when the observation stops, the rate often drops down to where it was before. This is a variant of the so-called Hawthorne effect: the behavior is altered (generally improved) just by the study itself.
Since hand hygiene is so simple to do and has such beneficial effects many facilities and organizations, including the World Health Organization and the Centers for Disease Control, have tried to push the compliance rate higher. An important component of this effort is to figure out ways to measure it without falling into the fallacies I noted above. How to do that?
Some studies have used closed circuit television to observe doctors and nurses. Others have used observers, but without letting the subjects know they are being watched. That seems a little bit sneaky, but it does produce more accurate numbers if you do it long enough so the caregivers forget it’s there. It is pretty labor-intensive, though, and therefore not well suited to ongoing quality control.
Recently I ran across a study of an interesting technology that monitors how often the hand sanitizer or soap dispenser in a patient’s room is activated. The notion was to validate that number against closed circuit TV observation to see if the simple number of activations of the dispenser corresponds to some expected number, on average, if everybody were doing good hand hygiene. The device works by sending an electronic signal to a remote server that keeps track. The researchers then developed an algorithm that linked activation of the dispenser signal with traffic in and out of the room as determined by video monitoring. The study compared washing inside the patient’s room with what is called the “in/out” method: caregivers are instructed to used a hand sanitizer dispenser beside the door to the patient’s room going in and coming out. They also had a phase of direct observation which, not surprisingly, increased compliance. The results were interesting:
This study validates the HOW2 Benchmark Study algorithm. It also documents a 36% deficit in hand hygiene opportunities using the In/Out method and a ~30% Hawthorne Effect due to direct observation. There is an extremely high correlation between actual video-taped hand hygiene compliance and the electronic monitoring system’s hand hygiene compliance index.
This all may seem a bit Big Brother-ish to people, but really, lives are at stake — hospital-acquired infections are a huge problem, and simple things like hand washing reduce them substantially. This is one way to get objective data to see if your educational efforts for the caregiver staff are getting anywhere. Although my hospital stresses the in/out protocol and has dispensers outside every patient room, I prefer to wash my hands in the sink at the patient’s bedside. For one thing, it’s been my routine for 45 years. For another, I think it is reassuring to families to see me wash my hands in front of them and their child.