NHLBI Working Group on Electronic Health Records:
Research Priorities to Improve Cardiopulmonary Health in Clinical Practice
The National Heart, Lung, and Blood Institute (NHLBI) convened a Working Group on March 10th and 11th, 2011. The purpose of the meeting was to identify high priority research topics in the use of the electronic health record (EHR) to support cardiopulmonary health and to recommend strategies that encourage expanding the capacity, use and standardization of EHRs in support of cardiopulmonary research. The Working Group consisted of experts in clinical care and research, legal and ethical issues, bioinformatics, biostatistics and EHR systems that currently support both clinical care and research activities.
This working group is responsive to NHLBI Strategic Plan Goals 1 and 2.
The working group discussion started with an introduction to NHLBI clinical research and trials. Four ongoing EHR-related research efforts were highlighted: the learning healthcare system from the Office of the National Coordinator for Health Information Technology (ONC), the NIH HMO Collaboratory and the NHLBI Cardiovascular Research Network, AHRQ’s DEcIDE (Developing Evidence to Inform Decisions about Effectiveness) Network Cardiovascular Consortium, and the FDA’s Mini-Sentinel program. The remainder of the meeting focused on different challenges and opportunities related to use of EHRs to improve cardiopulmonary research going forward. Each session identified top research priorities in that area, and the workshop concluded with a synthesis of key recommendations across all of the following topic areas:
- EHR data harmonization, standardization, and quality: How to share data.
- Sociology and mechanics of data sharing: Why share data?
- EHRs, registries, and observational studies: from cross-sectional to longitudinal.
- EHRs and randomized clinical trials.
- ELSI: ethical, legal and social issues with EHRs for research.
- EHR data analysis and cross-cutting opportunities and challenges.
The working group identified key priorities in three areas: Research, Policy, and Training.
- Clinical research networks allow researchers to identify and implement best practices to support collaboration, data sharing and validation studies across institutions. Such multidisciplinary research projects involve collaboration among researchers with key expertise (e.g., clinical, health services research, statistical, informatics, epidemiology, clinical trial, ELSI, etc.).
- The working group identified a need to develop more robust policy and technical solutions to address individual and institutional privacy and security concerns. Such solutions may include development of methods to improve informed consent processes in ways that increase participant understanding and participation, facilitate linkage of data to external sources (e.g., NDI, SSA, Medicare),or expand capacity for future research using EHRs.
- The panel identified a great need for more research on the development of quantitative methods for EHR-related research. Key topics include:
- Streamlining merging and analysis of individual-level data across EHR systems and health care settings while maintaining security of protected health information.
- Research using pre-existing data (e.g., registries, observational studies, completed interventional studies) for monitoring and improving cardiopulmonary health.
- Research developing effective ways to leverage EHRs within various study designs (e.g., cluster trials, pragmatic trials) to facilitate more efficient implementation of randomized clinical trials (e.g., administrative streamlining for IRB review and study contracting, real-time screening and recruitment, efficient data collection, outcome surveillance and confirmation).
- In order to establish the usefulness and reliability of EHR-based data for clinical care and research, the panel identified key research areas for systematic validation studies for EHR-based research. Key focus areas include:
- Research to develop and rigorously evaluate the effectiveness and safety of nationally available, standards-based, clinical decision support tools using the EHR for improving process measures and clinical outcomes in patients with NHLBI priority conditions.
- Research on how EHR data could be used to reduce bias and improve risk adjustment in comparing diagnosis and management strategies in observational studies.
- Assessing EHR-based interventions that aim to improve clinical care quality, care efficiency and transition coordination, support patient self-management and/or health disparities.
- A top concern raised by the working group was review of EHR-based research projects. They noted that scientific review panels that evaluate EHR-related grant applications need to include necessary and appropriate expertise, possibly through improved recruitment to standing study sections and/or establishment of special emphasis panel(s).
- The working group encouraged increased collaboration with other NIH Institutes, HHS agencies (e.g., CMS, FDA, AHRQ), and stakeholders to support development of EHRs that collect data in ways that can be used for multiple goals (e.g., clinical care, quality assurance and improvement, participation in registries).
- Partnering EHR-research projects with existing registries may encourage broader access to and use of existing and planned registry data collected as part of routine clinical and quality improvement efforts.
- The federal EHR standards and infrastructure being developed by the Office of the National Coordinator for Health Information Technology (ONC) may provide an opportunity to inform this national effort of the needs of NHLBI and its research community and expertise available to contribute to the staged development and implementation of meaningful use criteria.
- Cross-institutional collaborations may be facilitated by providing additional policy guidance and best practices to handle intellectual property issues, data use and transfers, and enhancing both individual and institutional privacy and security.
- Example documents for contracts and data use agreements may facilitate cross-institutional data sharing in NHLBI-sponsored EHR-based projects in a more timely and efficient way. Such a library of proven documents could be updated to reflect changing best practices in the research community over time.
- Expanded support for training of new and existing researchers to acquire advanced quantitative and methodological skills to effectively leverage EHR data for research. Expanded training is essential because of the unique challenges and characteristics of EHR data analysis, which are distinct from approaches best suited to traditional clinical research methodologies and approaches.
- The panel also noted that large-scale, collaborative EHR-based projects are an excellent opportunity to support innovative approaches to project leadership structures for investigators and institutions. Such projects should ideally include support for systematic training of young investigators in effective strategies for collaboration in multi-center EHR-based studies.
Working Group Participants
- Alan S. Go, MD, Kaiser Permanente of Northern California
- Robert Harrington, MD, FACC, FSCAI, Duke Clinical Research Institute
- Jessica Berg, JD, Case Western Reserve University
- Theresa Cullen, MD, Indian Health Service
- Lesley Curtis, PhD, Duke University School of Medicine
- Charles Friedman, PhD , Special Assistant to National Coordinator, ONC
- William E. Hammond, PhD, Duke University
- Adrian Hernandez, MD, Duke Clinical Research Institute
- Sharona Hoffman, JD, Case Western Reserve University School of Law
- Sheryl F. Kelsey, PhD, University of Pittsburgh
- David Magid, MD, MPH, Colorado Permanente Medical Group (CPMG)
- Daniel R. Masys, MD, Vanderbilt University School of Medicine
- Brent Muhlestein, MD, Medicine at University of Utah
- David C. Naftel, PhD, University of Alabama at Birmingham
- Lucila Ohno-Machado, MD, PhD, University of California San Diego
- Andy Podgurski, PhD, Case Western Reserve University
- Veronique L. Roger, MD, MPH, Mayo Clinic College of Medicine
- Justin Starren, MD, PhD, FACMI, Northwestern University
- James Troendle, PhD, NHLBI
- William Weintraub, MD, Christiana Care Health System
- Glen Bennett, PhD
- Jean Olson, MD MPH
- Dina Paltoo, PhD MPH
- Michael Twery, PhD
- Barbara Wells, PhD
- Lan-Hsiang Wang, PhD
Last Updated: October 2011