Wednesday, February 25, 2009

Evidence still not dominant driver of clinical practice in cardiology
By Maggie De Pano, DCRI Communications

Expert opinion—not evidence—is still the dominant driver of clinical practice in cardiology, a specialty reputed to be among the disciplines most firmly grounded in evidence. DCRI researchers found that many of the recommendations contained in the ACC/AHA Clinical Practice guidelines lack supporting evidence, and that the proportion of such recommendations is growing.

These findings were published February 24 in the Journal of the American Medical Association. Pierluigi Tricoci, MD, PhD, MHS led the study.

The American College of Cardiology (ACC) and the American Heart Association (AHA) have been releasing clinical practice guidelines for more than 20 years. These guidelines provide recommendations on cardiovascular disease care and cover diseases such as heart failure, interventional procedures such as percutaneous coronary intervention (PCI), and diagnostic procedures such as cardiac ECHOs. Guideline writers grade recommendations based on two factors: level of evidence and class.

Level of evidence combines expert consensus with an objective description of the existence and types of studies supporting a recommendation. Level A recommendations are based on evidence from multiple randomized trials or meta-analyses; Level Bs on evidence from a single randomized trial or on non-randomized studies; and Level Cs on expert opinion, case studies, or standard of care. Level C recommendations essentially lack supporting evidence.

Class of recommendation, meanwhile, indicates the strength of a recommendation based on guideline writers’ assessment of several factors, such as the relative importance of the risks and benefits identified by the evidence, as well as their synthesis of conflicting findings across multiple studies. Class I says that there is evidence and/or there is general agreement that a procedure or treatment is useful and effective; II that there is conflicting evidence and/or divergence of opinion; and III that there is lack of evidence and/or there is general agreement that a procedure or treatment is useless and ineffective.

Out of 53 guidelines issued by the ACC/AHA from 1984 to September 2008, the researchers studied a subset of 17 guidelines that were current as of September 2008. They found that among guidelines which were revised at least once, the total number of recommendations increased from 1,330 to 1,973, and that overall, this increase leaned toward more Class II recommendations.

They also found that among the 16 current guidelines reporting levels of evidence (cardiac ECHO does not), only 11 percent of the 2,711 recommendations were classified Level A, while 48 percent were classified Level C.

These findings indicate that the current system of generating research does not give caregivers and patients sufficient evidence-based information to help them consider the benefits and risks of therapy. The authors cited the scarcity of sponsors willing to fund studies designed to assess comparative effectiveness or routine clinical practice, as well as costly and protracted clinical trials, as likely contributors to the problem.

The authors concluded by highlighting the need to improve the process of writing guidelines to expand the evidence base from which clinical practice guidelines are derived, as well as encouraging interested parties in academia, industry, and government to collaborate in finding solutions.