Thursday, August 10, 2006

Diagnostic Uncertainty

Diagnostic Uncertainty. The phrase actually sends shivers down my spine. It is the big “I don’t know” of medicine. If there is uncertainty in the diagnostic process, then you can’t move on, you are stuck with the questions and the anxiety and the thing inside of you causing symptoms that could be a sign of anything. And, yet, diagnostic uncertainty can be a good thing too. It can be the yellow light that causes us to slow down and look both ways – to look for other symptoms and other diagnoses. Waiting a little longer for the answer, conducting that one extra test, just might be the right thing to do.

Healthcare providers should keep this lesson in mind, as well. Indeed, that was the finding of a group of researchers studying the treatment of pneumonia (see their article in the journal CHEST). It is generally agreed that pneumonia patients should be given antibiotic treatment within 4 hours of being seen at the hospital. In fact, hospitals are “graded” on their performance in this area and strive for 100% compliance with the practice. But here is where the diagnostic uncertainty comes in. The researchers found that there were some patients didn’t quite present as classic cases of pneumonia and the right thing to do in those cases is to delay treatment, even if it pushes the treatment past the 4 hour mark. Now my point here isn’t to discuss the validity of the study (the sample did seem a bit small) or the best practice, it is to remind healthcare providers (administrators mainly) that we can’t solely depend on best practice to guide medical decision-making. Even though healthcare is a business, medicine is a science, so they won’t always be in agreement over the right thing to do.

This is the second study that I have written about to question some of the best practices that hospitals are being asked to adopt. With the increased (and appropriate) focus on these practices, they are not likely to be the last. In the end, the industry is doing the right things – clinical guidelines, quality scorecards, pay for performance – and hopefully medical research will continue to inform it.