“This is the largest government study ever on its own people.”
Nancy Kass, Sc.D., a professor of bioethics and public health at the Johns Hopkins Bloomberg School of Public Health in Baltimore, was talking about the Precision Medicine Initiative, now called the All of Us Research Program.
Kass says she makes that bold statement deliberately and with humility, because she chairs the institutional review board (IRB) for the project, which aims to create a million-person cohort. “When the government does science on its own people, participant welfare and trust are essential and we may not get second chances,” she said. The promises made must be upheld, she added, and the National Institutes of Health (NIH) is making plenty of promises about issues such as diversity, transparency and ongoing engagement with communities.
Kass was speaking July 7 to an online meeting of the NIH Health Care Systems Research Collaboratory, which aims to improve the way clinical trials are conducted by creating a new infrastructure for collaborative research with healthcare systems.
The idea behind All of Us is that it provides resource efficiency because researchers do not have to recreate a cohort every time a new question comes up, which is how research is done now. Researchers can instead ask questions of a single huge cohort. The size also means people with rare conditions or circumstances are gathered in one place. “The ethical stakes are really high,” Kass said.
Citizens who participate agree to contribute their EHR and pharmacy data, as well as any health tracker data. They also take a minor physical exam and donate blood urine samples. The study will also look at health by Zip codes, too.
NIH is emphasizing that it wants the cohort to be diverse in many ways: geography, gender, race, ethnicity, sexual orientation, as well as health status and computer literacy. There are two methods of outreach and recruitment: One will involve recruitment of people in specific groups by provider organizations that have received grants to participate. The other is through direct volunteer recruitment of people who might see a sign about it on a bus or billboard.
Kass said the values NIH has espoused about the program are relevant to ethics and how the IRB sees its responsibilities. For instance, All of Us says participants will be partners in the program. “That is happening to some degree, but they are thinking further about ways that every site will be expected to uphold it,” she said. Another stated value is that trust will be earned through engagement and transparency. She noted that engagement means different things to different people, and that is why she is glad that three of the 10 members of the IRB are from community groups.
Another value is that participants will have access to information and data about themselves. How that will happen is unclear yet. “The devil is in the detail there,” she said.
All of Us has moved from pilot to the first phase of deployment and 500 people are enrolled. So far, the IRB has approved advertising material, a master protocol and is now approving recruitment sites. The NIH has funded 137 sites through 50 organizations. The IRB has approved six, with five approvals pending and four applications in process. “We know a lot more are in the pipeline,” she said. “We are helping to make sure the recruitment targets are being met at each stage to meet the goals.”
Kass praised the project’s innovative mobile consent process developed by John Wilbanks at Seattle-based Sage Bionetworks.
She noted that there are still many questions to work through, and that committees that have not even been created yet will propose policies and present them to the IRB for approval. For instance, one big issue to work through is whether participants entering the program are consenting to unforeseen future studies using their data or whether researchers would have to go back to them for consent each time their data is used. “NIH may set up some type of data access and use committee,” she said. “Our initial push has been getting the protocols approved and the cohort enrolled.”
One webinar participant noted that EHR data is included, and asked about data that insurers have on patients. Participants are told through the consent process that by participating they are agreeing to the study tapping into EHR but also insurance and pharmacy records. “The hope is to be expansive in the definition of electronic medical records, including insurance,” she said.
Rob Califf, M.D., the former FDA director, called into the webinar to ask about how the IRB would deal with the revised “Common Rule,” the federal policy for the protection of human subjects. Revisions were finalized in the January 19, 2017, edition of the Federal Register and subsequently subject to a “regulatory freeze” established by the Trump administration.
Kass replied that because the revisions haven’t been approved by the administration, the All of Us Research Program is still dealing with the former version. “We haven’t dealt with broad consent yet,” she said. “There will be attempts to make sure people have an understanding of what that means. I will say that the IRB has been conservative about a few things because of the responsibility it feels about precision medicine and upholding commitments on trust.”
Another question the IRB will have to address is how many other researchers will be able to access the data. Could citizen scientists gain access? The IRB needs to work through the safeguards in terms of qualifications of investigators as a mechanism to protect the data, Kass said. Yet another issue is which results should be returned to participants. Genetic results have not been handled that way previously, she noted. “There is a lot of work we have not started yet.”