The results of a recent observational study revealed a significant difference in start-up times for large cardiovascular trials in North America, highlighting changes needed for trials to become more efficient and feasible.1

This objective was to “measure the start-up times needed to reach prespecified milestones across sites in large cardiovascular RCTs [randomized clinical trials] in North America and to evaluate how these metrics vary by time and type of regulatory review process.”

Researchers studied nine trials conducted from July 13, 2004, to Feb. 1, 2017. The trials were coordinated by the Duke Clinical Research Institute.

Investigators found the overall site start-up time was almost nine months, although the top 10% of performing sites completed their start-up in less than four months.

The study authors cited problems for IRBs, including resources spent on the contractual process, protocols changing and requiring additional review, and centers with large portfolios deciding how to allocate resources.

“There was a significant but modest improvement in this metric when comparing early to contemporary trials, possibly related to the outpatient setting in which these later trials were conducted,” the researchers wrote.

“Contemporary trials more frequently used a central IRB and had faster times to regulatory approval when compared with earlier trials and sites using a local IRB. To our knowledge, this study for the first time quantitatively characterizes these performance metrics and provides insight into target areas for improvement.”1

“We were not terribly surprised by these results, having run clinical trials over the last 20 years in the United States,” says study co-author William Schuyler Jones, MD, an interventional cardiologist and associate professor at the Duke University School of Medicine and a member of the Duke Clinical Research Institute. “Ultimately, it placed a focus on the ability to start up and complete clinical studies, but it was an observational study.”

The data revealed using central IRBs can improve overall efficiency. The rationale for multiple sites reviewing clinical trial protocols through individual IRBs should be reconsidered. “Such multicenter clinical trial protocols are often reviewed by several independent parties, including academic leaders, data monitoring committees, and regulatory agencies,” the researchers wrote.

“It’s very clear that some of the top centers can start studies and enroll in studies in an efficient manner,” Jones says. “When we think of thousands of sites doing this work, central IRB is probably a key solution. But sites still can improve how they perform studies overall.”

“The use of central IRBs may enhance RCT start-up efficiency,” the authors concluded, “but more work is needed to ensure the timely implementation of a research protocol while protecting the interests of various stakeholders.”

A Harvard cardiologist found the results of this study “eye-opening.”

“I knew that IRB start-up times varied, but was surprised by the degree of variation,” says Deepak L. Bhatt, MD, MPH, professor at Harvard and executive director of Interventional Cardiovascular Programs at Brigham and Women’s Hospital. In a commentary on the study, Bhatt noted the cardiovascular clinical trial enterprise is in “disarray globally,” and the issues have been made more evident by the COVID-19 pandemic. “However, the crumbling clinical trial infrastructure, leading to inefficient trial execution, has been evident in the U.S. for a much longer time,” he wrote.2

For many reasons “spanning legal, regulatory, and cultural domains,” RCT enrollment in the United States remains “suboptimal,” with high costs per each patient enrolled, Bhatt noted. This potentially leads to a lack of generalizability of findings from international RCTs to patients in this country. “Over time, it will also lead to continued outsourcing of RCTs to other regions of the world and eventual loss of U.S. leadership in RCTs,” he wrote.2

In the most notable result, the start-up times of the top 10% of sites were less than half the other sites, Bhatt said. It is not a sustainable model for sites to take about three-quarters of a year from the time of study protocol delivery to enrolling their first patient. “No intelligent, fiscally responsible industry sponsor or not-for-profit organization would invest in such an enterprise if there were feasible alternatives available elsewhere in the world,” Bhatt wrote.2

Regulatory approval, contract execution, and site activation all are factors in slowing RCTs in the United States. “The parts that really slow things down are contracting and IRB approval,” Bhatt tells Medical Ethics Advisor.

Bhatt and the study authors agreed broader adoption of master service agreements between trial sponsors and sites could help streamline processes. “[H]aving master contracts between companies and institutions could help so that each trial contract doesn’t require the same wrangling back and forth about issues such as intellectual property,” he says.

It also might be more efficient to concentrate on high-performing sites. “This, of course, is dependent on these sites also being able to enroll a large number of patients, to provide thorough follow-up, and to maintain good adherence to the study protocol,” Bhatt wrote in his commentary. The more efficient sites also can share best practices with the slower sites.

The use of central IRBs also would “greatly enhance trial start up times in a way that would not compromise patient safety, and may even enhance it,” Bhatt tells Medical Ethics Advisor.

“As the saying among trialists goes, once you have seen one local IRB, you have seen one local IRB,” he wrote. “To a given trial protocol, there tend to be several idiosyncratic responses from individual well-intentioned IRBs, with one IRB finding a particular issue, and another finding a totally different one.”

IRB approval can be made more uniform and conducted by experts who have the time to devote to the task, but many hospitals do not have the resources to staff the IRB appropriately. Large academic centers might not want to cede authority to a central IRB.

“I understand that local IRBs work hard, try to do the right thing, and are proud of the work they do,” Bhatt tells Medical Ethics Advisor. “But the truth is that there is an enormous amount of variability in IRB skill sets and turnaround times. It is not clear to me that local IRBs generally add much in the way of patient protection for multicenter trials that have typically undergone several layers of regulatory and ethical board review. Central IRBs seem to handle multicenter trials much more efficiently, without any downside that I have seen to date.” He points out that local IRBs bring value in other situations, such as handling local single-site studies.

Bhatt says he has received an enormous amount of positive feedback about his editorial. “I can say that 100% of the clinical researchers I know agree with the points the authors made and that I made,” he notes.

Most of the reaction Jones received from his research contrasts with the findings from traditional studies to the ADAPTABLE (Aspirin Dosing: A Patient-Centric Trial Assessing Benefits and Long-Term Effectiveness) study. “It was a pragmatic, virtual clinical trial that used some of the newer methods,” he says.

In ADAPTABLE, patients with established atherosclerotic cardiovascular disease were randomly assigned to either 81 mg or 325 mg of aspirin per day. The study revealed substantial dose-switching to 81 mg of daily aspirin. The researchers did not find significant differences in cardiovascular events or major bleeding in patients assigned to 81 mg and those assigned to 325 mg.3

In this pragmatic trial, Jones and colleagues used “innovative and low-cost methods to simplify the identification, recruitment, and follow-up of patients.”

This included using algorithms in electronic health record data to identify eligible patients within the National Patient-Centered Clinical Research Network, providing web portal access for patients to give informed consent and obtain information on their aspirin regimen, and conducting trial visits via video chat or phone.

The author of a recent editorial called ADAPTABLE a “major achievement” because it provided a method of conducting efficient, low-cost trials in the United States that can be adapted quickly and used more widely. “This should allow many more clinical questions to be answered, with obvious benefits to healthcare consumers,” the author wrote.4

The observational study about cardiovascular trial start-up times highlights multiple key factors that are problematic for traditional clinical trials in the United States, Jones says.

“As clinical trials shift to a more patient-centered focus and a more pragmatic operational approach, we hope that people will be able to minimize start-up times and reduce the overall burden that’s required to conduct studies.”


  1. Goyal A, Schibler T, Alhanti B, et al. Assessment of North American clinical research site performance during the start-up of large cardiovascular clinical trials. JAMA Netw Open 2021;4:e2117963.
  2. Bhatt DL. Reconfiguring the cardiovascular clinical trial enterprise in the United States. JAMA Netw Open 2021;4:e2118176.
  3. Jones WS, Mulder H, Wruck LM, et al. Comparative effectiveness of aspirin dosing in cardiovascular disease. N Engl J Med 2021;384:1981-1990.
  4. Baigent C. Pragmatic trials — need for ADAPTABLE design. N Engl J Med 2021;384:2065-2066.