Penny Burgoon, Ph.D., director of the Office of Policy, Communications and Education at the National Center for Advancing Translational Sciences (NCATS), opened NCATS Day 2019 by welcoming the participants. She emphasized that the day was meant for conversations about how the participants viewed sharing data, not as a forum for discussing NIH’s data sharing policies. Dr. Burgoon said that different stakeholders brought different views to the discussion, and NCATS had invited a diverse panel of stakeholders, representing many points of view. She said that she expected it to be an interesting day.

Back to Agenda

Alice Chen Grady, M.D., a program officer in NCATS’ Office of Rare Diseases Research, introduced the first session, an interview with NCATS Director Christopher P. Austin, M.D., on translational science. She then introduced the moderator for the interview, Christina Hartman, M.P.H., senior director of advocacy at The Assistance Fund. Dr. Chen noted that Ms. Hartman is experienced in elevating the voices of patients, parents, scientists and clinicians to have a positive impact on health policy.

Ms. Hartman shared her interest in NCATS and her personal involvement with rare diseases. Below are highlights from Ms. Hartman’s interview of Dr. Austin.

What is translational science?

Translational science, Dr. Austin said, is something people understand even though they may not know the term. He explained that translation is the process of turning observations in the laboratory, clinic and community into interventions that improve the health of individuals and the public. He said that translation is the significant problem facing all biomedical research that led to the creation of NCATS; despite the explosion of innovation and knowledge in the last 35 years, the experience for the average patient in the clinic has remained about the same. He noted that despite the knowledge generated by fundamental science, most diseases still do not have any clinical treatment, and he added that this is the translational problem that translational science is meant to address. Translational science as a field, he said, seeks to find ways to make translation more efficient by understanding the scientific and operational principles underlying each step of the process.

Why does NCATS say that the patient is at the center of everything it does?

Ms. Hartman noted that the average timeline for diagnosis of a rare disease is five to seven years, so there is particular urgency for research among rare disease patients. She asked Dr. Austin how that urgency connects to the work NCATS does, including the robots he showed her during a tour of the NCATS labs. Dr. Austin said that the robots are exactly the kind of advance that responds to the research need Ms. Hartman mentioned. He noted that the robotic systems test thousands of drug compounds very quickly and can do in one week what a person would take 12 years to do. He added that the urgency and relevance that patients and patient families bring to the research process greatly help the push for change.

Dr. Austin noted that the decision to put the patient at the center of everything NCATS does was intentional. NCATS exists to help researchers develop treatments and cures for people. He related a story from his neurology residency about a patient with end-stage amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease), a rare disease for which there is no treatment. Dr. Austin said that watching the man’s agonizing death affected him and his career decisions profoundly. It made him want to find better ways to get cures to patients more quickly, he said, and the urgency from his encounter with that patient is what he keeps in his mind.

How does thinking about the connected nature of diseases fit into rare diseases research?

Ms. Hartman noted that when she first came into the world of rare diseases, she spoke with organizations that believed they were competing with one another for information and resources. She said that she believes a rising tide lifts all boats in science and that data collected to test a research question for one disease may be useful for research on other diseases and disorders. Dr. Austin suggested that rather than thinking of 7,000 rare diseases as individual diseases, people should think of them as 7,000 pieces of a giant jigsaw puzzle. The pattern still needs to be uncovered, he said, but as one piece is placed in relation to another, it becomes easier. He noted that this requires a different way of thinking about research, where research that benefits one disease also benefits other diseases.

What is NCATS doing to support data needs in rare diseases research?

Ms. Hartman said that getting a diagnosis of a rare disease allows patients to engage with that disease community, which can help piece together the natural history of disease progression. Having data on that history, Dr. Austin said, is critical to developing treatments, and NCATS is addressing data needs in several ways. As one example, Dr. Austin pointed to the Rare Diseases Clinical Research Network, which requires each consortium to have a natural history study. He also noted that NCATS is working to prevent a “tower of babel” problem, where different research platforms cannot “talk” to each other. He said that through the Rare Diseases Registry Program, NCATS is providing guidance and best practices for data standards and data elements to move toward a common language for data.

How does understanding rare diseases as a public health issue affect research?

Ms. Hartman said that millions of people are affected by rare diseases, but demonstrating the economic and social burden is difficult. Dr. Austin said that a fundamental problem is that the burdens of rare diseases are mostly invisible to the actuarial community, which analyzes financial risk, because only 5% of rare diseases actually have a specific diagnostic code, without which it is hard to connect a disease to its direct costs. Indirect costs, he noted, such as the loss of productivity for an individual and the cost of caregiving, are an additional, difficult challenge.

How can the conversations at NCATS Day help promote data sharing?

Ms. Hartman said that a challenge in data sharing is that there are deep divisions in perspectives, especially between healthy and sick people. Dr. Austin said that making the case for data sharing goes back to the connectedness of diseases and emphasized that finding those connections can be accelerated by sharing data in a common format. But to get everyone on board, he said, protections must be in place to prevent people from being worried about data misuse.

Dr. Austin concluded by saying that his hope for the day was to come to a common understanding of the value and potential of data sharing and of ways to maximize the benefits and minimize the risks of sharing data.

Back to Agenda

Elaine Collier, M.D., M.S., FACP, a senior advisor to the NCATS director, introduced Jennifer McCormick, Ph.D., M.P.P., an associate professor in the Department of Humanities at the Penn State College of Medicine. Dr. McCormick gave a brief presentation on data collection for research before joining the panel discussants. In her presentation, Dr. McCormick emphasized several points:

• Scientists collect and use a range of data from many different sources, including existing databases, animal models, tissue culture, electronic medical records, study participants and biospecimens.
• The process for collecting data from participants involves informed consent, which includes addressing a patient’s privacy and the security of their health information, information about what data will be shared and with whom, and information about how long data will be stored.
• There is a tension between having shorter, less detailed informed consents or longer, more detailed informed consents. Shorter informed consents offer only a broad description of the research but are relatively easy for researchers and participants to complete. Longer informed consents give participants more information and make the research more transparent, but they also are more of a burden for the participants and researchers completing them.
• Although people own their health data, they do not necessarily own the research data. Often research data are owned by the academic institution a researcher is affiliated with or by the funder or sponsor.
• Rights and responsibilities come with data ownership, including the rights to access and control the data and the responsibilities to protect and be a good steward of the data.
• There are many ways for data to be shared, including institutional agreements; large, web-based databases (including open access databases); consortia; and publication. Data sometimes are shared directly with participants, either as aggregate data from the group of participants or as individual research results.
• Motivations for sharing data include —
  • sharing individual research results with a participant for clinical use
  • sharing aggregate research data — combined data about a group — with participants to promote transparency and address participant interest
  • enabling research on “big data” in biomedical science with very large database

Dr. McCormick closed by noting that this was a transformative moment in terms of the ability to share data and the opportunities to put privacy protections in place.

Back to Agenda

Dr. Collier introduced Patricia Jones, Dr.P.H., M.P.H., a program director in the Division of Clinical Innovation at NCATS, who, in turn, introduced the panel discussion. Dr. Jones said that this was an important time in data sharing and for discussions about how to share data responsibly. She then introduced Marie A. Westbrook, Ph.D., of Inspire Concepts, LLC, who facilitated the panel discussion.

Panel participants had been asked to read two case studies in advance and come to the meeting prepared to discuss them. They were asked to consider several questions while reading the case studies:

• How was the participant data used and shared?
• What was the focus or purpose of the research?
• Who had control of the data?
• Were participants well informed?
• What important steps should have been considered?
• What other issues come to mind from these case studies?

The case studies and framing questions also were made available to all attendees before the meeting on the NCATS website.

Below are key points raised by panel participants during the discussion.

Preclinical Research Development

Matthew D. Hall, Ph.D., biology group leader, Division of Preclinical Innovation, NCATS, noted that there has been a change in the last 10 years in how data are shared, from thinking about how investigators should access data to thinking about how everyone should access data. He said that people are on the cusp of being able to do research using publicly available genetic databases and that such research would fall outside of the policy and regulatory framework researchers talk about and will bring a new dimension to managing data.

Clinical and Translational Research Teams: Research Investigators, Patient and Community Advocates, Biomedical Ethicists

Brendan Lee, M.D., Ph.D., Robert and Janice McNair Endowed Chair in Molecular and Human Genetics, professor and chairman of the Department of Molecular and Human Genetics at Baylor College of Medicine, said that his goal with patient engagement in research is not just involvement but leadership. If researchers think of patients as co-principal investigators, he said, it becomes natural that researchers would consult with patients before sharing or disseminating data in a publication. He added that he saw this as similar to the evolution in medicine from more paternalistic to more cooperative patient-physician relationships. Dr. Lee also noted concerns about access to bad data and the potential for machine learning to entrench biases. He pointed to data in electronic health records (EHRs) and noted that coding in EHRs is based on how things are paid for, which may not reflect the reality of the disease. Dr. Lee also noted that data utilization, not just data access, is an important issue for researchers and participants. Researchers are accustomed to data access being limited to a very small, specialized group of people, he said, noting that new tools will need to be developed that are appropriate to different levels of expertise in using data.

Louisa A. Stark, Ph.D., director, Community Collaboration and Engagement Team of the Utah Center for Clinical and Translational Science and director, Genetic Science Learning Center, University of Utah, said that involving nonscientist stakeholders throughout the research process can help inform how researchers approach data. But, she said, as scientists, they often do not know how to engage the community; it is not part of their training and often not part of the scientific community’s understanding of value. Dr. Stark said that NIH could help change that by including community engagement as a funding requirement. In her experience, education can lead to greater understanding of research and greater support for research. In addition to community and patient education, Dr. Stark emphasized the need to provide training for the research community on privacy, security and confidentiality.

Jennifer McCormick, Ph.D., M.P.P., associate professor, Department of Humanities, Penn State College of Medicine, said that it is important for researchers to be transparent with participants about how long data will exist and how it will be used. Dr. McCormick noted that engagement is a spectrum and said that scientists should be told about different ways to engage participants. Simply sending notifications and updates, she noted, is a form of engagement. She also noted that choosing to participate in research involves weighing risks and benefits and that people participating as healthy controls might see the risks of participating and receiving information back as higher than people in the rare diseases community. She said that researchers should take care in how information is disclosed and should respect that people may not want information back.

Eruera “Ed” Bryer Napia, Ph.D. candidate, M.S., program manager, Sacred Paths Youth Services and special projects director, Urban Indian Center of Salt Lake, emphasized the importance of clear communication between participants and researchers about expectations. Mr. Napia reflected on the fact that he can have conflicting feelings about participating in research. On the one hand, part of him thinks that if he could give a blood sample that would cure diabetes, he would want to give that sample. On the other hand, a part of him is concerned about the survival of his culture and of its spirituality, which would conflict with his desire to provide blood. Different concepts of life and death, he said, will affect how people choose to live and what and whether they are willing to donate to research.

Lori E. Crosby, Psy.D., professor, Division of Behavioral Medicine and Clinical Psychology at Cincinnati Children’s Hospital Medical Center and the Department of Pediatrics at the University of Cincinnati, College of Medicine and co-director, Cincinnati Center for Clinical and Translational Science’s Community Engagement Core and INNOVATIONS in Community Research and Program Evaluation, emphasized the importance of community education and establishing trust. She noted that her discussions of data sharing with community members, including adolescents and persons from underrepresented minority communities, address a range of topics and provide general education about the clinical research process, addressing what data are, how data are collected and how data are analyzed. With that background, she said, community members can then help researchers interpret data and answer questions that are important to the community. Dr. Crosby added that, particularly in the African American communities she often works with, some people do not trust researchers and believe that the researchers will not share the information collected during the study with the community. She said that it is important to identify and implement best practices around disseminating research and having stakeholders involved at all levels. Dr. Crosby noted that interest in receiving information may vary from not wanting to know any results to wanting to know everything. She said that these perspectives are essential to the discussion of responsible data sharing and, as researchers, they have a responsibility to work with communities. 

Tracy Smith Hart, CEO, Osteogenesis Imperfecta Foundation, pointed out that maintaining data is a fluid process because data change — people move, they change health statuses. Ms. Hart also said that patient groups are motivated to be involved in every part of the research process and want to be partners in research. She noted that they are good fundraisers and coalition builders and that they want to be at the table with researchers because it accelerates their mission. She said that a common complaint the Foundation hears from participants is that they never hear back from researchers. But, she said, participants also have a responsibility to communicate with researchers. Finally, she said that she thought most participants are hopeful about their participation and that their decision to participate comes from a place of making things better for people who come after them.

Pharmaceutical Industry

Robert Wright, Pharm.D., global head of scientific communications, Office of the Chief Medical Officer, Sanofi, said that industry had not been transparent in the past about data sharing. However, he said, this was changing with the development of the PhRMA Principles for Clinical Trial Data Sharing. He said that more companies are making a commitment to sharing aggregate data with patients and noted that Sanofi is part of a consortium that provides a forum for sharing data. Dr. Wright raised concerns about protecting patient privacy, especially when sharing patient-level data, even when it is anonymized. He also noted the importance of education about data, because sharing data widely would mean providing access to people who might not have the skills to know how to use the data, what the limitations of the data are, or why data created for one purpose cannot always be used for another. He noted that the discussion needs to go beyond issues related to creating the databases to addressing how to treat access to those databases by people outside of the classic research framework.

Federal Funders

Katherine Blizinsky, Ph.D., policy director, NIH All of Us Research Program, emphasized recognizing that there are individual needs to consider when people are deciding whether to take therapies or to receive information. Researchers, she said, should present options for receiving information in a way that does not pressure participants, and they should provide the support necessary for participants to interpret the results. She noted that this can be an easier issue to address when researchers are focused on one study with a specific question. She noted that the All of Us Research Program is generating a large cohort, which adds complexity, and could not ensure that a researcher would be able to reach back to the community from which the data originated. Scientists, she said, often forget that data are from actual people and that it is the responsibility of the government, as a funder of research, to set the tone for a shift in how the scientific community recognizes the human aspect of the data they collect.

Rubin Baskir, Ph.D., AAAS science and technology policy fellow, NIH All of Us Research Program, said that the decision to know or not know about genetic testing results is one that individuals make themselves, but it can affect a family or a community. Individuals who receive that information, he said, then have a burden of deciding what to do with it and the extent of their responsibility. He highlighted the need to look at where relationships between researchers and participants need to be facilitated in some way. This might take the form of training, community advisory boards or encouraging researchers to talk to community leaders. He noted that trust is fragile and that researchers have only one shot at doing research in a participatory way. He said that researchers should rise to the challenge and do research in a more relationship-oriented way.

Back to Agenda

Reflecting on the morning’s panel discussion on Exchanging Perspectives on Responsible Sharing of Data, NCATS Day breakout participants focused on three questions:

1. What points caught attendees’ attention?
2. What topics were not raised that attendees would like to see addressed in the future?
3. What data sharing best practices would attendees propose based on the panel discussion?

The highlights below reflect the themes that emerged in response to each question.

What points caught attendees’ attention?

• Data sharing practices can have implications for the individuals involved in the research, as well as for their families and communities for generations to come.
• Data can remain available for research indefinitely, which presents challenges for data security practices and informed consent.
• Encouraging researchers in academia to spend time on data sharing is a challenge, given the emphasis on publishing.
• Participants have different views on what information they want to share and what information they want to receive.
• Including participants and communities in research from planning through dissemination —
  • provides transparency and builds trust between researchers and the community
  • helps researchers understand how communities think about information sharing and develop processes and policies that take that into account
  • gives participants important context for discussions about data and an opportunity to learn about the research process

What topics were not raised that attendees would like to see addressed in the future?

• the role of institutional review boards
• participants’ health literacy and language access needs
• the growth of citizen science and use of public data sets
• considerations for compensating participants or communities
• responsible use of data from wearables and social media

What data sharing best practices would attendees propose based on the panel discussion?

• Informed consent: Use accessible language, multiple modalities (e.g., videos) and comprehension checks; include an explanation of the plan for data analysis, a clear description of benefits and risks, and a description of the long-term plan for oversight of the data.
• Building participant and community relationships: Have personnel with expertise in community engagement on the research team; include community partners as co-principal investigators; develop bidirectional learning opportunities about data sharing; provide advocates for vulnerable populations.
• Control of data: Maintain conditions for good data (e.g., security, encryption); bring a responsible end to participant data; inform all stakeholders of a data loss or breach.
• Access to data: Establish governance with clear ownership and ethical guidelines; develop incentives for researchers to share data; ensure data sharing policies have mechanisms to enforce compliance; provide skills training to nontraditional researchers.
• Information sharing and networking: Build collaboration around a common vision and understanding; develop shared definitions and documentation; create data sharing agreements that include training, conflict management, interoperability requirements, and processes and standards for quality control and assurance.

Back to Agenda

After Dr. Westbrook briefly recapped highlights from the breakout session, Dr. Burgoon thanked the planning group for their tremendous efforts.

Dr. Austin closed out the day by acknowledging the contributions of the NCATS Day participants. He thanked participants for suggesting ideas and problems for NCATS to work on.

Back to Agenda