Piloting an automated clinical trial eligibility surveillance and provider alert system based on artificial intelligence and standard data models.

BACKGROUND: To advance new therapies into clinical care, clinical trials must recruit enough participants. Yet, many trials fail to do so, leading to delays, early trial termination, and wasted resources. Under-enrolling trials make it impossible to draw conclusions about the efficacy of new therapies. An oft-cited reason for insufficient enrollment is lack of study team and provider awareness about patient eligibility. Automating clinical trial eligibility surveillance and study team and provider notification could offer a solution. METHODS: To address this need for an automated solution, we conducted an observational pilot study of our TAES (TriAl Eligibility Surveillance) system. We tested the hypothesis that an automated system based on natural language processing and machine learning algorithms could detect patients eligible for specific clinical trials by linking the information extracted from trial descriptions to the corresponding clinical information in the electronic health record (EHR). To evaluate the TAES information extraction and matching prototype (i.e., TAES prototype), we selected five open cardiovascular and cancer trials at the Medical University of South Carolina and created a new reference standard of 21,974 clinical text notes from a random selection of 400 patients (including at least 100 enrolled in the selected trials), with a small subset of 20 notes annotated in detail. We also developed a simple web interface for a new database that stores all trial eligibility criteria, corresponding clinical information, and trial-patient match characteristics using the Observational Medical Outcomes Partnership (OMOP) common data model. Finally, we investigated options for integrating an automated clinical trial eligibility system into the EHR and for notifying health care providers promptly of potential patient eligibility without interrupting their clinical workflow. RESULTS: Although the rapidly implemented TAES prototype achieved only moderate accuracy (recall up to 0.778; precision up to 1.000), it enabled us to assess options for integrating an automated system successfully into the clinical workflow at a healthcare system. CONCLUSIONS: Once optimized, the TAES system could exponentially enhance identification of patients potentially eligible for clinical trials, while simultaneously decreasing the burden on research teams of manual EHR review. Through timely notifications, it could also raise physician awareness of patient eligibility for clinical trials.

Communicating and disseminating research findings to study participants: Formative assessment of participant and researcher expectations and preferences.

INTRODUCTION: Translating research findings into practice requires understanding how to meet communication and dissemination needs and preferences of intended audiences including past research participants (PSPs) who want, but seldom receive, information on research findings during or after participating in research studies. Most researchers want to let others, including PSP, know about their findings but lack knowledge about how to effectively communicate findings to a lay audience. METHODS: We designed a two-phase, mixed methods pilot study to understand experiences, expectations, concerns, preferences, and capacities of researchers and PSP in two age groups (adolescents/young adults (AYA) or older adults) and to test communication prototypes for sharing, receiving, and using information on research study findings. PRINCIPAL RESULTS: PSP and researchers agreed that sharing study findings should happen and that doing so could improve participant recruitment and enrollment, use of research findings to improve health and health-care delivery, and build community support for research. Some differences and similarities in communication preferences and message format were identified between PSP groups, reinforcing the best practice of customizing communication channel and messaging. Researchers wanted specific training and/or time and resources to help them prepare messages in formats to meet PSP needs and preferences but were unaware of resources to help them do so. CONCLUSIONS: Our findings offer insight into how to engage both PSP and researchers in the design and use of strategies to share research findings and highlight the need to develop services and support for researchers as they aim to bridge this translational barrier.

Overcoming non-compliance with clinical trial registration and results reporting: One Institution's approach.

ClinicalTrials.gov is a web-based resource which provides the general public, healthcare professionals, patients, and caregivers access to privately and publicly supported clinical trials and trial results. The web site is maintained by the National Library of Medicine (NLM) at the National Institutes of Health (NIH) (ClinicalTrials.gov Background, 2018). The penalties for non-compliance with the legal obligations under FDAAA 801 (Food and Drug Administration Amendments Act of 2007) and the NIH requirements for registering and reporting results on studies within certain timeframes can result in large monetary fines and the withholding of federal funds (ClinicalTrials.gov FDAAA 801 and the Final Rule, 2019). Years following, in 2016, the Final Rule expanded upon the requirement with additional data elements for both registration and result submission records in accordance of FDAAA 801 (ClinicalTrials.gov FDAAA 801 and the Final Rule, 2019). The Medical University of South Carolina (MUSC), along with the institution's Office of Clinical Research and Regulatory Knowledge & Support group, identified issues affecting their own compliance rate with FDAAA 801 and the NIH and implemented several processes to overcome these challenges. In short, these processes included hiring a designated full-time ClinicalTrials.gov coordinator, implementing a workflow that identifies trials early in the IRB approval process requiring registration (without effecting study start up timelines), assisting researchers when navigating the registration and results reporting process through one-on-one consultations, Lunch and Learns, and disseminating new training tools as they become available. Over the next 12 months the results of this approach demonstrated a marked increase to 98% overall compliance with these federal regulations which may provide valuable guidance for other institutions working toward improved compliance rates.

Regulatory Support Improves Subsequent IRB Approval Rates in Studies Initially Deemed Not Ready for Review: A CTSA Institution's Experience.

We evaluated the impact of a regulatory support service (known as the Regulatory Knowledge and Support [RKS] program), part of the Medical University of South Carolina's Clinical and Translational Science Award, on the success of Institutional Review Board (IRB) applications that have previously been deemed by the IRB to be Not Ready for Review (NRR). At the time of this evaluation, 77 studies had been deemed NRR, 53 of which came from trainees and junior faculty. All the applications that received regulatory support either received IRB approval or were deemed to not be research, and therefore did not require IRB review. In all, 39.1% (n = 18) of the research teams who did not accept regulatory support successfully received IRB approval. Providing regulatory support, particularly to trainees and junior faculty, may be associated with better success in obtaining IRB approval as well as preventing the unnecessary submission of projects that are not research and would therefore not require IRB review or approval.

Providing Restricted Access to an Electronic Medical Record for Research Monitoring.

As hospital systems and healthcare institutions adopt electronic medical records (EMRs), this creates a new challenge in the normal conduct of clinical research. When protected health information (PHI) is stored in an EMR, there is inherent risk that general access to these systems for source verification purposes could allow research monitors to also have access to the PHI of non-study participants.

IRB reliance: An informatics approach.

Multi-site Institutional Review Board (IRB) review of clinical research projects is an important but complex and time-consuming activity that is hampered by disparate non-interoperable computer systems for management of IRB applications. This paper describes our work toward harmonizing the workflow and data model of IRB applications through the development of a software-as-a-service shared-IRB platform for five institutions in South Carolina. Several commonalities and differences were recognized across institutions and a core data model that included the data elements necessary for IRB applications across all institutions was identified. We extended and modified the system to support collaborative reviews of IRB proposals within routine workflows of participating IRBs. Overall about 80% of IRB application content was harmonized across all institutions, establishing the foundation for a streamlined cooperative review and reliance. Since going live in 2011, 49 applications that underwent cooperative reviews over a three year period were approved, with the majority involving 2 out of 5 institutions. We believe this effort will inform future work on a common IRB data model that will allow interoperability through a federated approach for sharing IRB reviews and decisions with the goal of promoting reliance across institutions in the translational research community at large.

Development and pilot testing of a video-assisted informed consent process.

The informed consent process for research has come under scrutiny, as consent documents are increasingly long and difficult to understand. Innovations are needed to improve comprehension in order to make the consent process truly informed. We report on the development and pilot testing of video clips that could be used during the consent process to better explain research procedures to potential participants. Based on input from researchers and community partners, 15 videos of common research procedures/concepts were produced. The utility of the videos was then tested by embedding them in mock-informed consent documents that were presented via an online electronic consent system designed for delivery via iPad. Three mock consents were developed, each containing five videos. All participants (n = 61) read both a paper version and the video-assisted iPad version of the same mock consent and were randomized to which format they reviewed first. Participants were given a competency quiz that posed specific questions about the information in the consent after reviewing the first consent document to which they were exposed. Most participants (78.7%) preferred the video-assisted format compared to paper (12.9%). Nearly all (96.7%) reported that the videos improved their understanding of the procedures described in the consent document; however, the comprehension of material did not significantly differ by consent format. Results suggest videos may be helpful in providing participants with information about study procedures in a way that is easy to understand. Additional testing of video consents for complex protocols and with subjects of lower literacy is warranted.

Quality assurance of research protocols conducted in the community: the National Institute on Drug Abuse Clinical Trials Network experience.

BACKGROUND: Quality assurance (QA) of clinical trials is essential to protect the welfare of trial participants and the integrity of the data collected. However, there is little detailed information available on specific procedures and outcomes of QA monitoring for clinical trials. PURPOSE: This article describes the experience of the National Institute on Drug Abuse's (NIDA) National Drug Abuse Treatment Clinical Trials Network (CTN) in devising and implementing a three-tiered QA model for rigorous multi-site randomized clinical trials implemented in community-based substance abuse treatment programs. The CTN QA model combined local and national resources and was developed to address the unique needs of clinical trial sites with limited research experience. METHODS: The authors reviewed internal records maintained by the sponsor, a coordinating site (Lead Nodes), and a local site detailing procedural development, training sessions, protocol violation monitoring, and site visit reporting. RESULTS: Between January 2001 and September 2005, the CTN implemented 21 protocols, of which 18 were randomized clinical trials, one was a quality improvement study and two were surveys. Approximately 160 community-based treatment programs participated in the 19 studies that were monitored, with a total of 6560 participants randomized across the sites. During this time 1937 QA site visits were reported across the three tiers of monitoring and the cost depended on the location of the sites and the salaries of the staff involved. One study reported 109 protocol violations (M = 15.6). Examples are presented to highlight training, protocol violation monitoring, site visit frequency and intensity and cost considerations. LIMITATIONS: : QA data from the entire network were not easily available for review as much of the data were not electronically accessible. The authors reviewed and discussed a representative sample of internal data from the studies and participating sites. CONCLUSIONS: The lessons learned from the CTN's experience include the need for balancing thoroughness with efficiency, monitoring early, assessing research staff abilities in order to judge the need for proactive, focused attention, providing targeted training sessions, and developing flexible tools. The CTN model can work for sponsors overseeing studies at sites with limited research experience that require more frequent, in-depth monitoring. We recommend that sponsors not develop a rigid monitoring approach, but work with the study principal investigators to determine the intensity of monitoring needed depending on trial complexity, the risks of the intervention(s), and the experience of the staff with clinical research. After careful evaluation, sponsors should then determine the best approach to site monitoring and what resources will be needed.

Cognitive-behavioral stress management for individuals with substance use disorders: a pilot study.

Stress-induced craving and stress reactivity may influence risk for substance use or relapse to use. Interventions designed to attenuate stress-induced craving and stress reactivity may serve as excellent adjuncts to more comprehensive treatment programs. The purpose of this study was to (1) tailor an existing, manualized, cognitive-behavioral stress management (CBSM) intervention for use in individuals with substance use disorders and (2) preliminarily evaluate the effects of the intervention using an experimental stress-induction paradigm. Twenty individuals were interviewed and then completed a psychological stress task, the Mental Arithmetic Task (MAT). After this, participants were assigned to either the CBSM intervention group or a nontreatment comparison group. Approximately 3 weeks later, participants completed a second MAT. In contrast to the comparison group, the CBSM group demonstrated significantly less stress-induced craving (p<.04) and stress (p<.02), and reported greater ability to resist urges to use (p<.02) after the second MAT. These findings are among the first to report on the use of an intervention to attenuate craving and stress reactivity among individuals with substance use disorders. Although preliminary, the findings suggest that systematic investigation of interventions specifically targeting stress management in individuals with substance use disorders should be undertaken.