Less than 10% of treatment recommendations for heart patients are based on high-quality evidence, according to the authors of a recent study.1 “We may need wholesale changes in the way we conduct clinical research,” says senior author Renato D. Lopes, MD, MHS, PhD.

More than a decade ago, researchers from Duke University examined the evidence supporting guideline recommendations in cardiology for the first time. They found that only 11% of recommendations in American College of Cardiology/American Heart Association (ACC/AHA) guidelines were supported by evidence from randomized controlled trials — the highest-quality level of evidence.

The researchers called for greater collaboration among investigators and funders in identifying key research questions, streamlined clinical trial methods, and more funding for clinical research.

“Over the past 10 years, some of these steps have been taken. But it is unclear how the evidence supporting guideline recommendations has changed,” says Alexander Fanaroff, MD, the study’s lead author.

Researchers analyzed 51 current cardiovascular guidelines — 25 from the European Society of Cardiology (ESC) and 26 from the ACC/AHA. Overall, 8.5% of ACC/AHA guideline recommendations and 14.3% of ESC guideline recommendations were supported by evidence from randomized controlled trials.

The proportion of recommendations supported by data from randomized controlled trials actually decreased from 2008. In looking at updated guidelines, the researchers found that fewer recommendations were supported by randomized controlled trials than in the prior versions.

“Despite the efforts of so many over the past 10 years, it seems we haven’t moved the needle on evidence generation,” says Fanaroff.

The incremental changes made in the last decade may not be enough, says Lopes, a professor of medicine at Duke University Medical Center in Durham, NC. More dramatic change is needed, he argues, including:

• Systems to use data generated in the course of routine clinical practice into randomized controlled trials;

• Relaxation of regulations on enrolling patients into clinical trials and monitoring during trials;

• A greater focus on funding collaborative, multicenter efforts to conduct critical clinical trials.

“It will also take a concerted effort to demonstrate to patients the importance of clinical trials,” says Lopes. A small proportion of patients participate in clinical trials currently.

“Many important questions don’t lend themselves to clinical trials as the clinical research ecosystem is currently constructed,” adds Lopes. Most studies are funded by the pharmaceutical and device industries. “But the medical research community should change the system so that these questions can be answered in a rapid fashion,” says Lopes. This will take a team effort from researchers, physicians, the pharmaceutical and device industries, hospitals, insurance payers, and patients.

Changing the structure of informed consent is one possibility. “Randomization at a level other than the patient, community consent, and other mechanisms could help generate evidence that helps many, many people,” he says.

However, research participants’ rights also need to be protected. Ethicists play an important role in this balancing act, says Lopes.

“An understanding of the lack of evidence underlying clinical guidelines and decision-making, and the benefit of generating this evidence, will help them best guide this process forward.”


1. Fanaroff AC, Califf RM, Windecker S, et al. Levels of evidence supporting American College of Cardiology/American Heart Association and European Society of Cardiology guidelines, 2008-2018. JAMA 2019; 321:1069-1080.


• Renato D. Lopes, MD, MHS, PhD, Professor of Medicine/Division of Cardiology, Duke University Medical Center/Duke Clinical Research Institute, Durham, NC. Phone: (919) 668-7993. Email: renato.lopes@duke.edu.