Workshop seeks to expand and improve data sharing
By Ernie Hood
Nearly 100 environmental health researchers, NIEHS program administrators, and other interested government officials gathered Feb. 6-7 in Research Triangle Park, N.C., to work together on the emerging practice known as data sharing.
Data sharing is well established in the genomics realm, with clear rules and guidelines in place for scientists to make their data broadly and quickly accessible to others. With environmental health science (EHS) data, however, the issues are more complex. That led NIEHS to issue a Request for Information (RFI) in June 2011, soliciting input and recommendations from the EHS research community, to use as the foundation for the Data Sharing Strategies for Environmental Health Science Research Workshop.
Why share environmental health science data?
In her opening remarks, NIEHS Division of Extramural Research and Training Director Gwen Collman, Ph.D., said there are solid reasons for NIEHS to work to expand EHS data sharing. “We at the Institute are quite interested in maximizing the NIH investments in the research we have funded over the years, to expand the opportunity for secondary analyses, for merging existing data resources, and for coordination between the various institutes’ missions, and, in general, to address some very common and challenging, complex disease/environmental exposure relationships,” she noted.
The meeting’s first keynote presentation by Bruce Lanphear, M.D., also addressed that question. Lanphear, a professor of children’s environmental health at Simon Fraser University in Vancouver, British Columbia, and a pioneer in research on the adverse effects of environmental lead exposures, explained that roughly 80 percent of the scientists in the audience had at some point relied on data collected by someone else.
“That’s one of the main reasons we should share data,” Lanphear said. “If we want our data to matter, if we want it to be useful, we should expect it to be public. We should be willing to share it.”
Broad themes and unique considerations
In presentations from experts and in discussions, workshop attendees addressed each of the major themes that emerged from the responses to the RFI, including protection of privacy and confidentiality, Institutional Review Board issues, legal and regulatory issues, NIH programmatic and logistical considerations, and computational challenges. Attendees also explored the many unique aspects of EHS data sharing (see text box) that were part of the driving force for convening the meeting, with an eye toward eventually shaping best practices and guidelines.
“We felt that there are particular issues with environmental health science data that are not there for gene-centric data,” said Kim McAllister, Ph.D., an NIEHS program administrator who co-organized the workshop. “So we want to make sure that any guidelines or recommendations we give our grantees would incorporate those unique considerations.”
One key concept that emerged during the workshop was the recognition that there is no single idea when it comes to data sharing. “It seems to me that we are talking about three different kinds of data sharing,” noted Howard Andrews, Ph.D., from the Columbia Center for Children’s Environmental Health.
“There’s data sharing that is voluntary and collaborative, there’s data sharing that is voluntary and non-collaborative,” Andrews said, “and, then there’s data sharing that is involuntary,” such as data sharing required by litigation or Freedom of Information Act requests. “There needs to be a framework that takes into account, in all three of those categories, a variety of aspects, such as adequacy of funding, adequacy of consent, conflict of interest issues, and more.”
McAllister said that the next step in the process is likely to be smaller follow-up workshops “to really flesh out some of these particular themes, just taking one of those areas and having some experts in that area in a think-tank sort of setting.” Ultimately, the goal is to delineate a broad NIEHS data-sharing strategy that will help overcome some of the current impediments to data sharing and foster collaborative, cross-disciplinary research.
(Ernie Hood is a contract writer for the NIEHS Office of Communications and Public Liaison.)
Unique considerations for the sharing of environmental health data
- The variation of environmental and biological measurements
- The potential to identify individuals based on the association of environmental exposures with geographical data
- The increased interest in return of individual or community-level environmental research results
- The regulatory implications of the use of environmental exposure and health data in developing U.S. national research policies
- The unique concerns of vulnerable populations who are disproportionately impacted by environmental exposures