EBM12: Future of evidence-based medicine is transforming
CHICAGO—The concepts of evidence-based medicine are being transformed as the healthcare system is being transformed, said Peter C. Wyer, MD, chair of the section on evidence-based healthcare at the New York Academy of Medicine and Columbia University Medical Center, speaking Oct. 4 at the CMIO Leadership Forum: Transforming Healthcare through Evidence-Based Medicine.
The question now is how IT can both mediate and enhance the whole framework of evidence-based practice as we move into a new information age, he said.
There is “fragmentation of what we would conveniently call evidence-based medicine.” The turf battles between the different streams of information haven’t helped us, he said. More evidence-based healthcare will embrace all dimensions–policy, guidelines and practice delivered to individual patients.
There are a lot of pluses to where we have been, Wyer said. Evidence-based medicine tightened the approach of looking at clinical trials, placing the emphasis on patient-important outcomes. To the credit of evidence-based medicine, addressing these issues of entrepreneurial or researcher bias, evidence-based medicine has brought about a heightened awareness of the liabilities in pharmaceutical-delivered product research and exposed the prevalence of things like reporting bias. There has been suppression of negative trials and use of composite endpoints to obscure outcomes. Evidence-based medicine is not just clinical trials. It’s a discipline with the best tools we have to protect ourselves from being played like accordions by the pharmaceutical industry.
Early limitations include oversimplified equations, such as therapy equals randomized trials and diagnosis equals performance accuracy. Another limitation is limited hierarchies. To date, the hierarchy includes randomized, controlled clinical trials, observational studies, pathophysiological investigations and practitioner opinion. This is not a legitimate hierarchy, Wyer said. Also, practitioner opinion is not a study design and it doesn’t belong in a list that includes controlled trials. Meanwhile, clinical judgment or skills and patient expertise inform the interpretation of all research studies and “pervade the entire spectrum of research.”
An ongoing challenge of evidence-based healthcare, Wyer said, is failure of dissemination which is “dramatic on all levels. Evidence-based medicine teaching does not affect practitioner behavior.” There is, in fact, an evidence transfer gap. “The fruits of research don’t make it into the world of clinical practice. Only a tiny trickle gets through from the research enterprise.”
Going forward, researchers and clinicians are “beginning to grapple with two different kinds of information.” He cited a review of the efficacy of nicotine replacement therapy. Researchers considered the results of 132 trials with 40,000 subjects. Even the studies that went out of the way to simulate real-world conditions, such as a subject going to a pharmacy for a therapy, favored nicotine replacement therapy. However, more recently, a large well done observational survey used rigorous methods to identify a potentially eligible set of subjects—people who had quit smoking. When looking at the association of nicotine replacement therapy and the likelihood of relapse, they found that if anything, the use of nicotine replacement therapy was associated with an increased likelihood of relapse. That could relate to the inherent bias of observational vs. randomized trials, Wyer said. The “discrepancy between the two kinds of information leads to a heightened level of knowledge about what to do.”
The new Institute of Medicine guidelines emphasize stakeholder involvement, patient involvement and transparency, Wyer said. Taking a patient-centered approach and starting with what patients are concerned with, want and need is becoming the basis of study design.
There are new approaches to clinical trials these days, such as pragmatic vs. explanatory trials; broader inclusion criteria; relaxed monitoring; active treatment to all patients; real-world conditions; and patient-centered outcomes.
The “whole thing comes to a head” when we start not looking at results in abstraction, Wyer said, which represents a significant change. “We haven’t talked about the actual context in which real patients are cared for by real practitioners.” That’s becoming more tightly defined as initiatives like accountable care begin to incorporate almost all of practice within some kind of organized framework.
To close the gap between research knowledge and clinical practice, “we see the different between push and pull,” he said. Push is the prevailing notion of how research knowledge gets into practice whereas a process for pulling research knowledge into the care setting is needed. “A pull process would have solved a lot of dissemination and implementation issues. You need both push and pull.”
New pull modalities include the EHR, administrative databases and practice-based research. “We’re at the dawn of the actual information age. We are at the beginning of an explosion of different kinds of information. I invite people to look at the EHR not just as source of pushing and decision support aimed at implementing means of research but a powerful means of pulling.”
In summary, Wyer offered the following objectives:
The question now is how IT can both mediate and enhance the whole framework of evidence-based practice as we move into a new information age, he said.
There is “fragmentation of what we would conveniently call evidence-based medicine.” The turf battles between the different streams of information haven’t helped us, he said. More evidence-based healthcare will embrace all dimensions–policy, guidelines and practice delivered to individual patients.
There are a lot of pluses to where we have been, Wyer said. Evidence-based medicine tightened the approach of looking at clinical trials, placing the emphasis on patient-important outcomes. To the credit of evidence-based medicine, addressing these issues of entrepreneurial or researcher bias, evidence-based medicine has brought about a heightened awareness of the liabilities in pharmaceutical-delivered product research and exposed the prevalence of things like reporting bias. There has been suppression of negative trials and use of composite endpoints to obscure outcomes. Evidence-based medicine is not just clinical trials. It’s a discipline with the best tools we have to protect ourselves from being played like accordions by the pharmaceutical industry.
Early limitations include oversimplified equations, such as therapy equals randomized trials and diagnosis equals performance accuracy. Another limitation is limited hierarchies. To date, the hierarchy includes randomized, controlled clinical trials, observational studies, pathophysiological investigations and practitioner opinion. This is not a legitimate hierarchy, Wyer said. Also, practitioner opinion is not a study design and it doesn’t belong in a list that includes controlled trials. Meanwhile, clinical judgment or skills and patient expertise inform the interpretation of all research studies and “pervade the entire spectrum of research.”
An ongoing challenge of evidence-based healthcare, Wyer said, is failure of dissemination which is “dramatic on all levels. Evidence-based medicine teaching does not affect practitioner behavior.” There is, in fact, an evidence transfer gap. “The fruits of research don’t make it into the world of clinical practice. Only a tiny trickle gets through from the research enterprise.”
Going forward, researchers and clinicians are “beginning to grapple with two different kinds of information.” He cited a review of the efficacy of nicotine replacement therapy. Researchers considered the results of 132 trials with 40,000 subjects. Even the studies that went out of the way to simulate real-world conditions, such as a subject going to a pharmacy for a therapy, favored nicotine replacement therapy. However, more recently, a large well done observational survey used rigorous methods to identify a potentially eligible set of subjects—people who had quit smoking. When looking at the association of nicotine replacement therapy and the likelihood of relapse, they found that if anything, the use of nicotine replacement therapy was associated with an increased likelihood of relapse. That could relate to the inherent bias of observational vs. randomized trials, Wyer said. The “discrepancy between the two kinds of information leads to a heightened level of knowledge about what to do.”
The new Institute of Medicine guidelines emphasize stakeholder involvement, patient involvement and transparency, Wyer said. Taking a patient-centered approach and starting with what patients are concerned with, want and need is becoming the basis of study design.
There are new approaches to clinical trials these days, such as pragmatic vs. explanatory trials; broader inclusion criteria; relaxed monitoring; active treatment to all patients; real-world conditions; and patient-centered outcomes.
The “whole thing comes to a head” when we start not looking at results in abstraction, Wyer said, which represents a significant change. “We haven’t talked about the actual context in which real patients are cared for by real practitioners.” That’s becoming more tightly defined as initiatives like accountable care begin to incorporate almost all of practice within some kind of organized framework.
To close the gap between research knowledge and clinical practice, “we see the different between push and pull,” he said. Push is the prevailing notion of how research knowledge gets into practice whereas a process for pulling research knowledge into the care setting is needed. “A pull process would have solved a lot of dissemination and implementation issues. You need both push and pull.”
New pull modalities include the EHR, administrative databases and practice-based research. “We’re at the dawn of the actual information age. We are at the beginning of an explosion of different kinds of information. I invite people to look at the EHR not just as source of pushing and decision support aimed at implementing means of research but a powerful means of pulling.”
In summary, Wyer offered the following objectives:
- We need a broad definition of evidence-based medicine to address the population system and patient issues.
- Traditional evidence-based medicine contributes methodological rigor and filtered electronic databases.
- Traditional evidence-based medicine reflects an incomplete model of dissemination and diffusion.
- Effective healthcare requires multiple kinds of evidence, including evidence generated from within practice settings.
- Quality is co-defined by methodological rigor and practical applicability and reality.
Beth Walsh,
Editor
Editor Beth earned a bachelor’s degree in journalism and master’s in health communication. She has worked in hospital, academic and publishing settings over the past 20 years. Beth joined TriMed in 2005, as editor of CMIO and Clinical Innovation + Technology. When not covering all things related to health IT, she spends time with her husband and three children.