Data Sharing Certifications (NIH)

The NIH requested institutional certification to share my study data in a repository, and/or to approve my grant proposal's data sharing plan. What now?

OSP normally signs these for Emory researchers, but first they ask the IRB for our input.

The IRB weighs in on whether the data sharing is ethical, and if it complies with any applicable policies and regualtions.

  • Ethical: whether subjects would reasonably understand that their data might be shared in this manner, based on the informed consent form/discussion
  • Compliant:
    • If consent is waived, does the sharing align with the Belmont Principles and human subjects regulations? This may depend on what identifiers are involved.
    • If the study was NIH funded, did the informed consent align with the NIH policy in place at the time the data or samples were collected?

Genomic Data Sharing is a subset of data sharing that falls under a special NIH policy (the "Genomic Data Sharing" policy). only large-scale genomic analyses falls under the policy - as defined by the NIH. The NCI has an excellent web site explaning the policy: Click Here

The IRB needs four things as part of a request for institutional certification:

  1. All applicable IRB numbers for studies that collected (or will collect) the data to be shared, and the samples that generated the data, if applicable
  2. If large-scale genomic data sharing is involved (see website above), and the data is generated under an NIH-funded study, include the date range during which the samples were collected (or state that all samples will be collected propspectively)
  3. Copies of all consent form versions used to collect the samples/data to be shared (or the draft consent form, for a new study)
  4. A pre-filled copy of the institutional certification, with the investigator's draft answers to the questions about how data use should be limited (if at all). Answers should be based on the consent form(s) the subjects signed.

If the data sharing certification is for a multi-site study under a different PI, who used your existing data from an Emory study, please explain the situtation in your email request, including the study titles/PI's involved, along with the above information.

In rare cases, the IRB may need to request an amendment to an existing study to revise the scope of HIPAA/consent waivers. Or, the IRB may may determine that data sharing is not consistent with the consent forms the subjects signed. If this occurs for a study that already took place, you will will need to discuss next steps with your NIH representative.