IRB Guidance and Best Practice on Depositing Data Into Large Databases

Increasingly the NIH and other federal funding sources insist that data, especially genomic data, generated in research are put into large nationally accessible databases. If a repository is funded by NIH (such as dbGAP or dbSNP), there are very particular specifications that must be met before researchers can deposit data. Per the requirements of the NIH, the IRB needs to be able to certify certain things. In order to do so, researchers need to have specific information in the protocol regarding the data sharing plan and language in the consent form describing what the researcher is doing with the data, the risks unique to depositing large amounts of genetic data, and what it means to have one’s data in such a database. While this article highlights work funded by NIH and other federal sources, the IRB supports a single standard approach and regardless of funding, suggests all projects meet the criteria listed below if depositing data into repositories. Links to the NIH Policy are at the bottom of this article.

If researchers plan to deposit data into large databases, a clear statement regarding deposits must be in the consent form and best practice is to include an option for subjects to agree/not agree to this use of their research data in the consent form.

Please read this information carefully and keep the suggested language available when crafting consent forms.  It is recommended that you modify your current project and consent forms if your vision is/was to share data and you don’t currently have that process approved in your project.      

Data sharing plans include data type, data repositories to which data will be submitted, appropriate uses of the data and limitations on the future use or an exception to submission which explains why the Institutional Certification criteria cannot be met and must describe an alternative mechanism for data sharing. Please see The NIH Guidance for Investigators in Developing Genomic Data Sharing Plans for more information.

If you are planning on depositing data into a large-scale repository, NIH requires the IRB must assure certain criteria are met:

  1. The protocol for the collection of data is consistent with 45 CFR 46;
  2. Data submission and subsequent data sharing for research purposes are consistent with the informed consent of study participants from whom the data were obtained;
  3. Consideration was given to risks to individual participants and their families associated with data submitted to NIH-designated data repositories and subsequent sharing;
  4. To the extent relevant and possible, consideration was given to risks to groups or populations associated with submitting data to NIH-data repositories and subsequent sharing; and
  5. The investigator’s plan for de-identifying datasets is consistent with the standards outlined in the GDS (Genomic Data Sharing) Policy.

In order for the IRB to make the assurance, it needs to review the protocol, data sharing plan details, and informed consent document.

The protocol needs to include information about the data sharing, future use and deposits to repositories, in addition to statements about the potential privacy risks to the subjects and their families (as well as to their kin group or population, if a specific population or group is being targeted or investigated).

The informed consent document must include detailed information about future use including what data will be shared, with whom and for what purposes. In addition to detailed future use language, the following language must be used:

 

IRB Approved Language for Consent Form(s):

Under Description

To do more powerful research, it is helpful for researchers to share information they get from studying human samples. They do this by putting it into one or more scientific databases, where it is stored along with information from other studies. Researchers can then study the combined information to learn even more about health and disease. If you agree to take part in this study, some of your genetic and health information might be placed into one or more scientific databases. There are many different kinds of scientific databases; some are maintained by [institution], some are maintained by the federal government, and some are maintained by private companies. For example, the National Institutes of Health (an agency of the federal government) maintains a database called “dbGaP.” A researcher who wants to study the information must apply to the database. Different databases may have different ways of reviewing such requests. Researchers with an approved study may be able to see and use your information, along with that from many other people. Your name and other information that could directly identify you (such as address or social security number) will never be placed into a scientific database. However, because your genetic information is unique to you, there is a small chance that someone could trace it back to you. The risk of this happening is very small, but may grow in the future. Researchers will always have a duty to protect your privacy and to keep your information confidential.


Under Risks

Group Risks
Although we will not give researchers your name, we will give them basic information such as your race, ethnic group, and sex. This information helps researchers learn whether the factors that lead to health problems are the same in different groups of people. It is possible that such findings could one day help people of the same race, ethnic group, or sex as you. However, they could also be used to support harmful stereotypes or even promote discrimination.

Privacy Risks
Your name and other information that could directly identify you (such as address or social security number) will never be placed into a scientific database. However, because your genetic information is unique to you, there is a small chance that someone could trace it back to you. The risk of this happening is very small, but may grow in the future.  Since the database includes genetic information, a break in security may also pose a potential risk to blood relatives as well as yourself. For example, it could be used to make it harder for you (or a relative) to get or keep a job or insurance.  If your private information was misused it is possible you would also experience other harms, such as stress, anxiety, stigmatization, or embarrassment from revealing information about your family relationships, ethnic heritage, or health conditions.

There is a Federal law called the Genetic Information Nondiscrimination Act (GINA). In general, this law makes it illegal for health insurance companies, group health plans, and most large employers to discriminate against you based on your genetic information.  However, it does not protect you against discrimination by companies that sell life insurance, disability insurance, or long-term care insurance.

For more information on the GDS policy, data use limitations, writing data sharing plans, and more visit  https://gds.nih.gov/06researchers1.html.

For Guidance on Future Use and Large-Scale repository considerations from the PPHS, please refer to PPHS Guidance.