HCLF: Clancy on improving evidence-based medicine

CHICAGO—“You could shoot holes in what we’re doing” related to evidence-based medicine and quality assessment, said Carolyn M. Clancy, MD, former director of the Agency for Healthcare Research & Quality, speaking at the Healthcare Leadership Forum on Nov. 14.

“There has been an explosion in evidence-based medicine,” she said, but a more interesting way to think about it is “if evidence-based medicine is the answer then what is the question? The ultimate question is how do we apply the best of science to patient care taking into account people’s needs, preferences and beliefs.”

The questions are pretty straightforward, she said: Which treatments work for which patients and what is the trade off? “Ultimately most people would like to provide the best possible care to their patients but how do you get this information out in a way that’s an easy, routine part of the day? It’s not so much any part of the routine of what clinicians do. You’re making it up as you go along. Starbucks has more built-in error checking than most of medicine.”

Until recently, there were few tools to get from evidence to practice, she noted. “This is not a technical problem. We don’t have GPS for medicine that makes the right thing the easy thing to do.” The challenge is how to organize information so vendors can help deliver information in the right place at the right time.

For the large part of medical practice, “we have no idea what works but we pay for it anyway,” she said. Unfortunately, developing measures of quality is difficult because there is a “choke hold around data capture. Evidence by itself doesn’t have that much meaning. Evidence based on good science and endorsed by one or more professional organizations has a much different meaning in the medical politics sphere. We know that public reporting of quality performance is associated with improvements in care but it’s not a silver bullet.”

Local data are required to reduce variation and improve quality, Clancy said. National benchmarks are important but strong local coalitions are needed to engage with the public and those who pay for care. “This is a shared enterprise. We can get to a place where we’re using the best of evidence to get to good quality measures but we have a delivery system ill-equipped to actually incorporate it.”

Health IT will make a lot of this easier, she said, but “evidence is not neutral and completely unbiased. It exists in a context. Stakeholder engagement in that process is really important.”

Looking forward, she said we need qualitative information because variation in quality cannot be solely attribute to the fact that some doctors spend more time reading research articles. “There are many nonclinical factors that affect decision making and we need to understand a lot more about that at a microsystem level.”

Clancy said she predicts changes in practice. “Are we up to this task? I’m wildly optimistic that we are but I think it will be rough sailing at times.”

The Healthcare Leadership Forum was sponsored by Clinical Key and presented by Clinical Innovation + Technology.