Proposed consent process gives volunteers options for personal data
By Ruth SoRelle, M.P.H.
When are data truly anonymous – unable to be traced back to the individuals from whom they were derived?
The question is not academic when applied to genetic databases, said two Baylor College of Medicine researchers in a recent essay that appears in the journal Science.
As DNA technologies improve and the number of publicly available genetic databases mushroom, information that people once thought could not be linked to individual patients or volunteers may not be as anonymous as was once thought, said Amy L. McGuire, Ph.D., J.D., and Richard Gibbs, Ph.D.
In a report titled "No longer de-identified," McGuire, an assistant professor in the BCM Center for Biomedical Ethics and Health Policy, and Gibbs, director of the BCM Human Genome Sequencing Center, predict that preventing identification of an individual's identity directly from DNA data will present a more frequent challenge as the number of public databases grows. One recent estimate shows that a person in such a database could be identified from only 75 of the millions of one letter variations called SNPs (single nucleotide polymorphisms) that differentiate one person's genetic blueprint from that of another.
Technology growing
"The technology and data are each growing at a tremendous pace," said Gibbs "so we must stay ahead of what that means to those who volunteer for genetic studies."
"The technology with regard to identification is moving rapidly," said McGuire. "In the future, the possibility of identification from information in the database is going to be greater."
Informing people of those risks and setting up systems to give them more control about how the information is to be released is important, she said.
"My concern is the possibility of undermining the public trust," said McGuire. Individuals might be less willing to participate in genetic research if they are not adequately informed about the possibility that they could be identified in some way from a public database.
Three-tier process
To ensure that people who participate in genetic studies are well informed about future privacy risks, and to allow them to consider different options about how their data may be further distributed, the authors propose a new three-tiered consent process. In this way the participants can control where data are displayed and with whom it is shared.
"We want to continue to improve the public trust," said McGuire. "If the possibility arises that a person could be identified against their wishes using data from a study, it could be harmful to genetic research in the future."
The consent is different from most because it gives research subjects options instead of telling they can participate or not.
"We are suggesting a stratified consent process to give people the option of choosing what they want done with their data. We want people to participate in research and we would like a broad variety of populations of people to participate," she said.
The stratified system would give them an opportunity to take part even if they don't want their data released publicly.
