We must rely on the best evidence we have.
The practice of evidence-based medicine is the integration of individual clinical expertise with the best available external clinical evidence from systematic research, and further integrating this with the patient's values and expectations.
This is not cookbook medicine; the evidence needs extrapolation to the patient's condition and context. Evidence-based medicine involves judgment under uncertainty and is influenced by biases, heuristics, and framing. Yet it's preferable to "conventional wisdom," which uses individual clinical experience rather than patients as the denominator. Conventional wisdom tends to focus on high-tech treatments and relies on simple literature searches (conducted without systematic analysis) that may miss more than half of the most rigorous types of evaluations.
Psychotherapy in double-blind, randomized trials has shown efficacy for a variety of conditions. Medications show efficacy for a variety of conditions. That's not the issue. The issues are: What tools do we use to evaluate a treatment? How does it work? Who does it work for? When do you apply it?
When psychiatrists use antidepressants, I assume they think they know why they're using them. We all are using some evidence for making choices among medications. Is the evidence good enough? It is true that the level of evidence that we have in psychiatry is not what we would like it to be.
We can't wait for perfect evidence. We always make choices--judgment under uncertainty. We would like perfect evidence, but we have to use the evidence that we have to make a choice. If you don't want to use evidence to guide your decisions, you can do whatever you want to do, to whomever you want to do it. That's not good practice.
Those of us in psychiatry should not feel as if the level of evidence with which we deal is worse than in other fields of medicine. It's not. The level of evidence that we get isn't perfect, but it's reasonable. The question is how to frame things. Is the glass half full or half empty? We have reasonable evidence for many of the things we do, but not for all of them.
Any time you get a new piece of evidence, you really have to look at what you're doing and ask, first, is this evidence right for my patient? Evidence-based medicine is aimed at teaching you how to evaluate the evidence and apply it. If the trials are done wrong and not relevant to the patients you're seeing, you're going to have to weigh the evidence in a different way. If it's done on exactly the type of patient you see, then it's generalizable.
A lot of the evidence in medicine, not just psychiatry, falls in that category. It is rare to get evidence on exactly how to treat somebody who is 39 years old and has high blood pressure, diabetes, and depression. Instead, we collect evidence from X, Y, and Z, and apply it to our patients. It is an art and it is a science. The science is collecting the evidence and looking at it.
At some point, our field will have to focus on the real outcomes--functional outcomes--which we have not studied well to date. What do we really mean when we say somebody got better?
For now, however, we rely on the best evidence available. Better and more relevant evidence is needed. While we wait for that, learning the principles of how to use evidence in the best interests of our patients is critical.
DR. KRISHNAN is professor and chair of psychiatry at Duke University, Durham, N.C.
K. RANGA R. KRISHNAN, M. …