Some claim that Direct Observation (D.O.) is the best method to both (A) measure, and (B) change hand hygiene behavior. We assert that D.O. can do one or the other, but not both. (And it does neither particularly effectively, which we’ll get to later.)
First, let’s look at using D.O. to measure hand hygiene behavior. Let’s put aside most of the problems with D.O., such as observer bias, recall bias, small sample size, inability to see into rooms, etc., and focus on the impossible-to-correct Hawthorne Effect.
In this scenario, in order to gather reliable data, the person doing the tracking (a “secret shopper”) must remain anonymous. If a clinician knows he’s being watched, he’s much more likely to clean his hands, skewing the data. So the only way to capture accurate data is to remain secret.
But if the secret shopper is able to remain anonymous, it’s impossible for her to change behavior. Sure, you can provide reporting at the end of the month, but for numerous reasons this is not an effective way to change behavior.
In some hospitals, the person with the clipboard will attempt to change behavior directly. She may witness a clinician forget to clean his hands, and then approach him to remind him in the moment. While potentially effective in changing behavior, the secret shopper is no longer secret. With her cover blown, the Hawthorne Effect takes over, and clinicians will sanitize when they see the not-so-secret shopper. Several studies have shown that this artificial inflation only occurs when the observer is present and immediately goes away when the observer leaves.
Direct observation, by its very nature, cannot simultaneously both measure and change hand hygiene behavior.
You might ask yourself, is it better to have marginally better data about a problem that you can’t fix (focusing D.O. on measurement) … or to potentially fix a problem, knowing that the data you have is totally worthless (focusing D.O. on changing behavior)?
We realize that hospitals have to do something to keep the accreditation agencies out of their hair, but that’s about all that Direct Observation is good for. Of course, electronic hand hygiene performance systems like ours solve both problems. We offer comprehensive, objective measurement combined with in-the-moment behavior change. No need to make tough choices.