Using evidence in clinical decision making is like trying to navigate your way to a destination, let’s say a museum, in an unfamiliar city. There are various different ways you might try to figure out how to get to the museum. You might ask a random person on the street, you might ask your friend, you might consult an expert like a tour guide, or you might use an internet-based mapping program like Google maps. Though you may have different degrees of confidence in each of these approaches, essentially what you are concerned with is finding your destination. Personally I would prefer to use Google maps as it gives clear directions and is very precise and so we would expect to find the museum easily.
 
Imagine now that, using Google maps, were unable to find the museum. Maybe we take a wrong turn, or perhaps there are two streets with same name. Given the reliability of Google maps (it has directed us well to other destinations) and our diligent map reading, we might be tempted to believe it is likely that the museum does not exist, forget about it, and move on to landmarks that are easier to locate. Regardless, the fact that we couldn't locate the museum doesn't prove that it doesn’t exist; it simply means that we haven’t found it yet!

Let’s say now that we ask a random person on the street if they know where the museum is. Having lived in the city for some time they inform us that its name has been changed, or it has been relocated or maybe it is simply difficult to find. Eventually, despite the vague directions we received from the stranger and the occasional wrong turn we do eventually find the museum. It really did exist after all! True we are not as confident as we might have been with Google maps that it is the same museum, nor has its exact location been verified by satellite photographs, but we really only cared about finding the museum to satisfy our sightseeing needs. Although it would be nice to be able find our way back next time…

The purpose of research then is reproducibility: how can we be sure that we get to the museum most of the time. The value of publishing positive studies is to let others know a route that we have found that gets us there. And the value of publishing negative studies ought to be to alert other explorers that we have tried a particular route and it wasn’t helpful and so not to waste their time going down that road.

The difficulty is that we often put too much faith in evidence-based medicine. The world of medicine is far from being successfully mapped. We often turn away from treatments because “there is no evidence” meaning Natural Standard or BMJ Clinical Review categorizes them with “insufficient evidence”. The truth is that, as complementary and alternative care providers, that is where we live. And so we must rely on evidence whose methodology is less than robust. However, it is a grave error to say that this makes the treatments less effective, or the museums less remarkable, it simply means Google has catching up to do.  In the words of Larry David: “Whatever works!”