Creating safe and effective treatments, diagnostic tools and medical devices that improve human health requires successful testing of those interventions in humans. Researchers nationwide face common barriers in recruiting (or accruing) enough participants for clinical trials. The inability to identify and recruit the right number and type of people to participate often:
- Makes clinical trials slow and more costly;
- Limits the validity of trial results and, in turn, researchers’ ability to apply the findings broadly to the general population; and
- Stops a trial prematurely or prevents it from taking place at all.
In fact, a recent analysis of more than 7,500 Phase II and III cancer trials registered on ClinicalTrials.gov between 2005 and 2011 found that 20 percent were never completed. The most common reason: inability to recruit participants.
One promising way to identify clinical research participants is to access information contained in electronic health records (EHRs) across the country. This solution is not without problems: For example, in addition to legal limits and privacy concerns, typically, the medical data are not linked across institutions, and EHRs frequently do not “talk” to each other easily because they use different terminology for the same information.
CTSA Program Solutions
NCATS’ Clinical and Translational Science Awards (CTSA) program supports efforts to solve system-wide translational research problems in part by developing and implementing ways to improve the success of U.S. clinical trials. One initiative, CTSA Accrual to Clinical Trials (CTSA ACT), was launched recently to do just that by developing a nationwide network of sites that share EHR data. Building on existing platforms and operating models to create a “federated” network with common standards, data terminology and shared resources, CTSA ACT investigators are focused on:
- Data harmonization (using the same term for the same type of data) across EHR platforms;
- Technical needs assessment and implementation;
- Regulatory approaches to ensure compliance with protocols for data access and participant contact; and
- Governance development to establish proper agreements among institutions.
“We are approaching the challenge of identifying and enrolling patients by linking EHRs to identify potential participants who meet study criteria,” said Steven E. Reis, M.D., a professor and associate vice chancellor for clinical research at the University of Pittsburgh, and director of the University of Pittsburgh Clinical & Translational Science Institute. He added that the ACT team created standard categories and terminology for demographic and clinical visit data as well as for medications and laboratory results.
“Recruiting participants into clinical trials is a critical success factor in our ability to bring more treatments to more patients more quickly,” said Petra Kaufmann, M.D., M.Sc., NCATS Division of Clinical Innovation director. “The CTSA ACT investigators are tackling this challenge using innovative tools not only to find out where potential participants are, but they also are beginning to look at better ways to then connect participants with research opportunities.”
The CTSA ACT technology platform has two major components:
- Informatics for Integrating Biology and the Bedside (i2b2), a software package that converts raw records into de-identified and searchable participant information stored in a central database.
- Shared Health Research Informatics Network (SHRINE), a search engine for i2b2. With appropriate agreements in place, an ACT investigator can use SHRINE to search de-identified records using a customized set of participant criteria. SHRINE queries all network institutions and provides an approximate number of participants at each site who meet the criteria.
The i2b2/SHRINE technologies are open source (available to anyone) and widely used, for example in NIH-funded studies at Harvard Catalyst, the university’s Clinical and Translational Science Institute. This enabled the ACT team to get the first sites up and running quickly. Because these technologies are just two of the solutions in use across CTSA hubs, one objective is to identify and integrate hubs that do not use i2b2 into the ACT network.
So far, ACT researchers have tested the technology at 13 CTSA hubs. All of these sites have obtained approval from their IRBs, and, to date, eight additional CTSA hubs have been selected for inclusion in the project moving forward.
Participant Screening and Recruitment
CTSA ACT will enable a qualified investigator to quickly determine, approximately how many participants might be available for a study and where they are located, all in a de-identified way. Once a scientist knows how many potential participants meet the criteria for a trial, the next step is to find out who they are and to contact them to discuss whether they would consent to participating in the trial and for additional screening. Any personally identifiable information must be shared in a legal and ethical manner; therefore, CTSA ACT will implement an approach that has overcome these challenges.
“We must protect privacy while advancing research, and we are actively discussing those issues,” said Gary S. Firestein, M.D., professor and dean and associate vice chancellor of translational medicine at UC San Diego. “We are sensitive to the fact that patients may be uncomfortable with the notion of someone in a different health system looking through their records. That’s why this work is done with de-identified data.”
One solution is for each site to have a designated local investigator who will serve as an “honest broker” or “data concierge.” This individual sorts through patient records to identify participants in that region and to figure out the right way to contact them. After a contact request is approved, normal IRB oversight and the local investigator ensure that participant privacy is protected.
CTSA ACT leaders also are working to ensure their efforts complement a similar Patient-Centered Outcomes Research Institute (PCORI) nationwide network under development, called PCORnet.
“As CTSA ACT and PCORnet take shape, we are aiming for synergy,” explained Robert D. Toto, M.D., professor and associate dean for translational science at the University of Texas Southwestern. “The goal is to ensure that we develop methodologies in concert with PCORI to make the ACT and PCORnet data interoperable for future initiatives.”
Existing Resource Collaborations
While unique in the number of institutions that will be joined in a cooperative network, CTSA ACT is building on a strong foundation of existing CTSA resources and models. One CTSA-supported federated network also using the i2b2/SHRINE technology is the University of California Research eXchange (UC ReX), a secure, Web-based system that enables UC investigators to identify potential study participants using data from patient care activities at the five UC medical campuses: Davis, Irvine, Los Angeles, San Diego and San Francisco. CTSA investigators are integrating UC ReX into CTSA ACT, and many of the best practices established in UC ReX’s development will inform ACT’s continued growth.
“Other CTSAs have developed informatics platforms and established participating sites, so we can build on that foundation,” Firestein said.
The ACT team also is tapping into the knowledge of CSTA-supported investigators who have developed a variety of patient registries aimed at improving trial accrual; some registries have recruited more than 10,000 participants. Frontiers: The Heartland Institute for Clinical and Translational Research is another CTSA hub with an established federated network, called HERON (Healthcare Enterprise Repository for Ontological Narration). Using the i2b2 technology as the foundation, HERON enables access to de-identified EHRs by faculty across the campuses in the University of Kansas system. The Frontiers network includes partners at the Kansas City University of Medicine and Biosciences, 10 hospitals and health systems in Kansas and Missouri, and 14 community partners in both states.
HERON has a patient registry with more than 30,000 participants. One approach used by Frontiers is to ask patients during office visits if they are interested in clinical research; patients can sign a brief consent form that authorizes research-related contact and inclusion of their medical information in a registry. Another is setting up a Web-based registry, such as the CTSA-supported ResearchMatch, for interested patients and healthy volunteers.
Both UC ReX and HERON are in early development, but investigators are collecting and analyzing data to determine the effectiveness of these networks in improving trial accrual. Examples from both suggest that using technology to streamline and improve the accrual process works. UC ReX enabled UC’s participation in a large, multisite comparative-effectiveness study — “Comparing Options for Management: Patient-Centered Results for Uterine Fibroids,” funded by PCORI and the Agency for Healthcare Research and Quality — by demonstrating in an efficient way that UC had enough qualified participants. Without UC ReX, the process had been lengthy and often inaccurate. With similar goals, Frontiers has made steadily increasing use of HERON, launched in 2012; in its first two years, researchers used the technology to enroll 150 participants in clinical trials. This number is expected to grow as investigators become used to the new process and the benefits it can provide.
Expanding CTSA Support
In addition to and building on these ongoing efforts, NCATS plans to fund two Recruitment Innovation Centers (RICs) in fiscal year 2016 through the CTSA program. The program will focus on improving accrual into multisite clinical trials through innovative informatics and technology-driven approaches as well as developing ethics and policy frameworks and improved ways to connect participants to trials.
“The RICs will support new ways to assess the feasibility of recruitment and selection of research sites for multisite studies while implementing strong privacy safeguards and sufficient permissions for data access,” Kaufmann said. “They will also help us find better ways to connect participants with research opportunities and collectively share our goal to enable evidence-based understanding of what works best in trial participant recruitment.”
Posted February 2015