Evidence-Based Medicine Experts Define the State of the Art

At the close of the CMIO Leadership Forum: Transforming Healthcare Through Evidence-based Medicine last fall in Chicago, Clinical Innovation + Technology invited five luminaries to build on the abundant discussions and dive deeper into the quality of the evidence in medicine, the lack of evidence in many cases, the struggle to engage physicians and patients, and strategies for integrating evidence-based medicine (EBM) into clinical practice. Will the adoption of EBM reach 90 percent by 2020? Should evidence-based medicine be expanded to evidence-based healthcare? Will we see the rise of EBM as a medical specialty? Read on to find out.

The moderator of the discussion was Michael S. Blum, MD, Chief Medical Information Officer, University of California San Francisco Medical Center, San Francisco, while the participants included:  

Andrew Auerbach, MD, MPH, Professor of Medicine in Residence, University of California San Francisco and Director of Research, Division of Hospital Medicine; Hospitalist, University of California Medical Center, San Francisco

Sameer Badlani, MD, FACP, Chief Medical Information Officer, Assistant Professor, Department of Medicine, The University of Chicago Medicine and Biological Sciences, Chicago

Jonathan Teich, MD, Chief Medical Information Officer, Health Sciences Division, Elsevier and Emergency Physician, Brigham & Women’s Hospital and Professor, Harvard Medical School, Boston; and,

Peter C. Wyer, MD, Chair of the Section on Evidence Based Health Care at the New York Academy of Medicine and Associate Clinical Professor of Medicine at Columbia University College of Physicians & Surgeons, Emergency Medicine Department, Columbia University Medical Center, New York, N.Y.

Blum: We have spent the last two days talking about evidence-based medicine. One of the main topics we touched on is the quality of the evidence. The Institute of Medicine (IOM) got a lot of this started. The IOM has upped the bar and is pushing us to integrate evidence-based medicine into our practices. Recently they predicted that by 2020, 90 percent of clinical decisions will be supported by accurate, timely and up-to-date clinical information, and will reflect the best available evidence. Our discussions indicated that maybe 20 percent of decision-making is currently evidence-based. What will the reality be less than eight years from now?  

Auerbach: I don’t think the next eight years is going to be any different than the last 30 or so. The breadth of the data almost never covers every clinical situation you are going to encounter in practice. What’s interesting in the emergence of EHR and EMR predictive modeling is the ability to use crowd-based evidence as a potential guide for what you are seeing at the bedside. We have spent some time [at the CMIO Leadership Forum] on how data contained in the medical record can help you understand how the last 20 patients with this disease did in your institution. That’s an interesting and unique feature of the EMR world. EMR has certainly become the glue that holds together these learning organizations.

Wyer: The goal of 90 percent [by 2020] is probably unrealistic. But that doesn’t mean healthcare shouldn’t continue the unrelenting march to integrate evidence. I think [the IOM’s] purpose is not so much to predict the future, but to set some goals and agendas and create some objectives which help steer us. I don’t think we are going to get to 90 percent evidence-based decision making with externally generated research along the lines of what we have been used to. The only way we are going to come close is if we recognize there are different kinds of evidence and that different category of evidence is most important, which is what Andrew just pointed out. It’s practice-based research, and the EMR is a critically important venue for collecting that.

Teich: The amount of decisions that are supported by evidence is a product of the amount of evidence that’s out there and the ability of people to see the evidence, pay attention to it and use it. There is going to be a certain amount of growth in the amount of evidence that’s out there. It may not get from 20 percent of all the space to 90 percent, but it will grow. I am hopeful there will be more chances to put that information in front of clinicians at the right time.
Blum: What does practice-based learning look like in the EHR and in practice?

Wyer: The EHR is a potentially powerful vehicle for gathering a much broader suite of information from practice. It transforms what we used to think we were talking about in terms of a practitioner’s awareness of his or her own experience. The EMR makes it possible for us to retrieve practice-based information. We actually can record experience. That itself is potentially transformative because we have converted something that was purely experiential into something that actually can be looked at objectively and thought about.

Teich: I concentrate on clinical decision support, which is really the delivery of all these things in to the right person at the right time. There is a lot of growth going on in terms of how are trying to bring that information to people’s attention. On the flipside, does that also mean that we can then have the ability to use more data to generate more information to go into the mix? Certainly. My concern is that may be level 3 on the priority of what people are thinking about right now with EHR, where level 1 is meaningful use and documentation and so on, and level 2—I wish it were 1.5—is clinical decision support. Then perhaps level 3, because it’s new, is being able to do the learning exercise. The question really is will people make it a priority.

Blum: There was a U.S. News & World Report article in which an orthopedic surgeon wrote that evidence-based medicine is becoming a specialty in its own right. Are we going to need a specialist in evidence-based medicine? Or is it something that every provider needs to have an understanding and ability around evidence-based medicine in his or her toolkit?

Badlani: I strongly disagree that evidence-based medicine should be its own separate field.  That begets the notion that I can be a really good cardiologist, and somebody else should be an evidence-based physician. I think evidence-based medicine needs to become part of our toolkit. You can be a super expert at it or a novice at it, but it still has to be part of what we learn as doctors.  There was a slide presented at the CMIO Leadership Forum showing what you needed to be a great doctor in 1977 and what do you need in 2012. Today, being familiar with the basic principles of evidence-based medicine is absolutely paramount.

Teich: It’s what we need. We also saw here that there is thousands and thousands of times too much evidence than one person can process. I know in my work, I rely on other people sometimes to tell me the evidence. Sometimes I try and get it myself; sometimes I listen to others. There are always going to be people who can be your evidence shop, can be your medical decision-making center of excellence. That may not be an ABMS medical specialty, but there will always be some people who spend more of their time doing it than others.

Auerbach: [Evidence-based medicine] needs to be part of the student curriculum. It needs to be a core competency for next generation [of physicians] who will, by the way, be graduating in 2020.

Blum: Peter, you said at the conference you don’t really like the term “evidence-based medicine,” but you favor “evidence-based healthcare.” You were really pushing on that concept of a learning environment and practice-based information.

Wyer: Medicine of any kind is not practiced in a vacuum, it’s practiced in organized settings. The direction we are on is going to expand that principle. ACOs are going to be bringing together folks who were previously working in the historical solo practitioner model. They are actually going to be part of a system through that process. So one has to look at issues such as what is knowledge in an organization and capacity of an organization to generate knowledge and to respond to innovation generated both from within the organization and externally. This actually redefines the whole framework of what we are talking about when we are talking about evidence.

Blum: The topic came up several times: monitoring what’s going on after you put [EBM] systems in place, learning from them and changing them. It really points out that these projects don’t just go in and the implementation isn’t just done, that you are always having to evolve.

Badlani: Even if you take all the academic institutions or all community hospitals, everybody is at a different level of the number of available resources, number of available individuals who are interested in this kind of work, and then the bandwidth of the local institution. But if you think about the common theme amongst all of them, we all have to report out on quality measures. We have an obligation to our payers and the federal government to report out multiple quality measures.
One way to make this easier and reduce the table stakes of participating in this is if we had a standard way of defining [common conditions such as] what is quality for DVT prophylaxis and diabetic care, and drive evidence-based medicine through that. That allows my community hospital or my small practice to have only one type of denominator, one type of numerator to worry about, and we could make a significant change across the institution.

Wyer: We heard about the notion of shared decision making as an alternative to “standard of care.” I think that’s somewhat crucial. It’s crucial to provide the flexibility that allows for ongoing innovation. The idea that we can just pin down the quality approach to a particular disease entity, even something as seemingly straightforward as DVT prophylaxis, is probably an illusion. The whole point of this is evolving knowledge that creates its own dimension of uncertainty. We need an approach that actually can ride with the tide of evolving knowledge.