View all hidden authors and organizations

In January 2021, the Northwell Health hospital system in New York launched a clinical trial to see whether the over-the-counter drug famotidine (also known as Pepcid) can reduce the severity of COVID-19 in symptomatic patients who do not need to be hospitalized. This randomized trial was started based on anecdotal reports and clinical studies that showed that Pepcid benefits COVID-19 patients (1). But the trial has a twist: it is completely virtual; there is no need to visit the research site (2). In fact, this is the first completely virtual clinical trial in the health system. Northwell leaders say this will not be the last.

"COVID has accelerated this process, and it works better," said Christina Brennan, vice president of clinical research at Northwell Feinstein Institute of Medical Research. Brennan said that the initial design of the trial required an on-site visit, and the trial leader began to switch after realizing that potential participants would prefer to rest at home. Brennan said that the virtual model allowed Northwell to recruit more diverse participants than other trials, possibly because on-site visits were not an obstacle. The medication (Pepcid or placebo) was mailed to the participants, and all laboratory draws were done in their homes. The trial was conducted by Northwell in cooperation with the Cold Spring Harbor Laboratory in New York State, and recruitment was completed in September 2021. Data analysis is in progress.

Smart watches are one of the technologies that can facilitate large-scale decentralized experiments. Patients monitor their health in daily life. Image source: Shutterstock/Andrey_Popov.

Northwell is not alone in direct mail—this practice has become common during the pandemic. "It's basically everywhere. In all our oral treatment trials, we keep patients at home," said Keith Flaherty of Massachusetts General Hospital in Boston. Flaherty said that the first priority in the early stages of a pandemic is to maintain access to experimental treatments, including those who have already participated in clinical trials.

In addition to direct mailing, some clinicians also allow participants to sign consent forms remotely rather than on-site, and researchers are remotely monitoring people’s progress during the trial [see Figure 1 (3)]. For example, this might involve using a smart watch to measure someone's heart rate or oxygen saturation level, or telemedicine visits via video, or reporting participants' responses to research treatments over the phone.

These changes together constitute a "decentralized clinical trial." Proponents say this approach allows faster and more diverse participation in research, while skeptics warn that conducting clinical trials in an uncontrolled environment may lead to data errors and flawed conclusions. Regardless, the COVID-19 pandemic has triggered a debate about the role and consequences of these unconventional studies.

Like traditional trials, decentralized trials can assess the impact of new drugs or lifestyle changes-for example, exercising more often. It can be completely remote, such as Northwell Pepcid research, or it can combine face-to-face and remote activities.

According to Craig Lipset who led Pfizer's first fully remote test in 2011, the idea of ​​decentralized testing can be traced back to the early 2000s (4). "And many of the ways we bundled together to call decentralized dates take longer than that, such as using electronic diaries," adds Lipset, who is the co-chair of a consulting company that aims to accelerate the adoption of decentralized trials. Lipset pointed to the feasibility study of the dispersion test in 2003 (5) and the fully dispersed anxiety and insomnia treatment study in 2005 (6).

But Lipset said that COVID-19 has finally fundamentally changed the tip of the needle. According to McKinsey, a consulting firm in New York, shortly before the pandemic, fewer than half of pharmaceutical company representatives or representatives of organizations that contracted pharmaceutical companies to conduct clinical trials expected remote trials to become a major part of their portfolios. A year later, when McKinsey re-examined this issue, the figure was 100% (7).

"Everyone said,'Oh, we will do this in the next five to ten years. It's really hard,'" Andy Kravos, CEO of HumanFirst, a digital health consulting firm in San Francisco, California. Andy Coravos) said. "Suddenly COVID strikes, and it's not that difficult."

Coravos has long been an advocate of decentralized clinical trials. When she wrote about the prospects in 2018, Kravos emphasized that less than 5% of the U.S. population participates in clinical research, and that more than 70% of the population are more than two hours away from academic medical centers. In addition, Coravos said, clinical trial sites are man-made environments that do not reflect the daily reality of how people live with diseases or health challenges. In theory, networked sensors such as smart watches, wireless blood pressure cuffs or home sleep monitors can bridge this gap (8).

The Heartline trial was launched in February 2020 just before the pandemic started, and it demonstrated the value of a decentralized approach (9). Due to the decentralized model, trial investigators successfully continued to recruit throughout the pandemic without changing their research protocols. Recruitment is ongoing, and the researchers hope to recruit 150,000 participants 65 years of age or older.

Heartline's goal is to understand whether early warnings of irregular heartbeat patterns and stroke warning signals (called atrial fibrillation (AFib)) can stimulate heart health behaviors and reduce the risk of stroke or death. All Heartline participants can access the Heartline app for iPhone, which reminds and advises people to practice heart health. Some participants have an Apple Watch, which can remind them of possible AFib instances. Heartline leaders wanted to assess whether the group that received these alerts was healthier than the group that did not receive these alerts. This was measured by the different rates of death or stroke in the two groups over time.

In traditional trials, people would only learn about potential AFib regularly during visits to the research site. Research leader C. Michael Gibson said that the dispersion model revealed previously invisible examples of AFib. It also allows clinicians to “collect serious results such as deaths and strokes,” adds Gibson, a cardiologist at Harvard University in Cambridge, Massachusetts. In other words, Gibson does believe that tracking more "subtle endpoints", such as fluctuations in the patient's vital capacity, will continue to require visits to the clinic. As this research is ongoing, it is not clear whether smartwatch alarms will actually improve health outcomes.

As doctors, researchers, and patients try to adapt to the COVID-19 pandemic, many trials have shifted to a decentralized format. Distributed trials place patients at the center of the study rather than the location of the clinical trial, and conduct consultation, consent, and testing remotely. Image source: Reprinted from reference. 3. Approved by AACR.

When conducting decentralized experiments, logistics can be challenging. Ken Getz of the Tufts Center for Drug Development and Research in Boston, Massachusetts predicted in the June issue of Applied Clinical Trials, “As we enter the post-pandemic world, layoffs and more conservative applications of adaptive and innovative practices” (10) . Getz points out that survey data shows that people conducting such experiments usually spend "excessive" time learning new software or coordinating different activities in remote areas.

Then there is the core task of any experiment: data collection. "I do have some concerns, mainly related to the completeness and effectiveness of the data collection," said Mary McDermott, a professor of medicine at Northwestern University in Chicago, Illinois, who treats and studies vascular diseases, which can be unbearable for the elderly. People walk easily.

McDermott points out, for example, the six-minute walk test, which measures the response of patients with peripheral artery disease to exercise strategies designed to increase walking ability. McDermott said the correct way to perform this test to ensure that the results are reliable and comparable is always to maintain the same surface along the 100-foot-long corridor; people walk back and forth as much as possible for six minutes. McDermott pointed out that at home, people may test on the carpet first, and then test on the hard floor, otherwise there will be no 100 feet of barrier-free space at the beginning. All these variables are controlled at the study site.

It is always worrying that ensuring patient privacy presents a special challenge for decentralized trials. Without strong privacy protections to specify how often and for what purpose data collected from health sensors can be used, insurance companies may discriminate against people whose smart watches show breathing difficulties or frequent heartbeats. As with any online information, hacking of such health data is another problem. "I really think we need some version of the genetic information non-discrimination bill to connect sensors and digital data," Kolavos said.

Part of the reason for these trials is to increase the diversity of participants. However, some public health experts believe that only through the active outreach of clinical trial leaders will diversity increase significantly. “Many low-income people, many people of color, do not have access to medical services in large academic research centers, so they don’t even know that clinical trials are happening,” said Harvard doctor Bisola Ojikutu, who was also involved in the research. The executive director of the center. Boston Public Health Commission.

"I do have some concerns, mainly related to the completeness and effectiveness of the data collection."

Ojikutu said that the person in charge of the trial should go to the community to explain the existence of their clinical trial and how to participate. And since many decentralized trials will have online components, Ojikutu recommends setting a budget for the data plan and any necessary equipment to ensure that no one cannot participate in the trial because they cannot access the Internet. In addition, to ensure that confusing applications or interfaces do not hinder participation, Ojikutu recommends that trial designers include technical literacy guidance so that all participants can participate equally.

Challenges aside, decentralized experiments seem likely to continue, at least in some form. The Cure 2.0 bill recently proposed in Congress requires FDA commissioners to provide guidance on how to regulate the regulation of decentralized trials, including how to evaluate digitally collected health data (11). Cures 2.0 may be submitted to Congress for deliberation before the end of the year.

Such a framework is essential to promote the widespread adoption of decentralized experiments. Flaherty said that compared with trial participants or clinicians, pharmaceutical companies are the most cautious about moving to decentralized trials because they worry that any new treatments that use this cutting-edge approach will not eventually be approved by the FDA. Even advocates of decentralized trials are proceeding cautiously. The Heartline trial does not involve any drugs. As part of clinical trials during the pandemic, Flaherty only began to mail drugs to patients through the emergency authorization provided by the FDA.

At the same time, the debate over whether decentralized experiments is wise continues. "People don't want to travel. I'm stronger than ever because it makes more sense at home," Kravos said. But others encourage clinicians to proceed with caution. "There are all kinds of things," McDermott pointed out, "this may interfere with the way the data is collected."

November 24, 2021: The article body has been updated.

Released under PNAS license.

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.