WPF advises National Institutes of Health re: Genomic Data Sharing Policy

WPF filed comments to the National Institutes of Health (NIH) regarding its Genomic Data Sharing Policy. WPF recognizes that NIH is attempting to create a reasonable standard. Nonetheless, WPF urged NIH to better “future proof” its genomic data sharing proposal, and in particular requested NIH to look further into key areas, including:

  • Limits of  deidentification regarding genetic data,
  • Impacts of “genomic big data,”
  • Secondary uses of genomic data,
  • Difficulties relating to providing access to genomic data for unregulated research purposes, and
  • New components to include in Data Sharing Agreements.

From the comments:

“We are not far from the time when relying on deidentification as a means of privacy protection for genetic data will be impossible as a practical matter. In the past, the dissemination of genetic data was only a trickle by today’s standards, and past levels of computing power facilitated deidentification as a powerful tool for privacy protection. However, that time is passing by, and rapidly so in the area regarding genetic / genomic data.

Today, increased computing power (particularly machine learning techniques) combined with the wide availability of genomic and non-genomic data sets from the public and private sector has advanced data analysis to the point that deidentification does not have the same utility as it once did. These capacities, combined with the much broader dissemination of genetic data both within and outside of HIPAA-covered entities, has created the availability of “genetic big data.”

WPF’s comments noted that “genomic big data” is particularly difficult to deidentify. This is a simple statement on its face, yet behind it lurks an entire sea-change that creates new problems.

WPF has done a lot of work over the years regarding genomic privacy. We are linking to this newest work, and a selection of related work, below.

Related documents: