Skip to main content
U.S. flag

An official website of the United States government

Return to Search

NIA-Specific Funding Policies

Guidance for NIA grantees on data sharing, funding line and cost control policies, research on human subjects, and NIA-specific guidance on particular grant mechanisms

Final

Issued by: National Institutes of Health (NIH)

https://www.nia.nih.gov/research/grants-funding/nia-specific-funding-policies

NIA-Specific Funding Policies

Policies Concerning Particular Grant Mechanisms

  • Program Project (P01) Policies and Guidelines
  • Ranking of NIA-reviewed P01, U01, U19, and R01 applications
  • NIA procedures for submitting applications for support of scientific meetings (R13, U13)
  • NIA Review of Complex, Large Applications (e.g., P01, U19): Beginning with applications submitted for the September 25, 2014 deadline, NIA will no longer require or allow investigators to be available on call during the review meeting for any applications. However, for program projects (P01) and other large and complex applications, requesting $1 million or more in direct costs in any year, when these applications are reviewed singly, NIA will retain the existing policy of soliciting questions from reviewers to submit to investigators at least one week prior to the review meeting. This policy does not apply to applications reviewed by standing committees or submitted in response to RFAs.

NIA Funding Line and Cost Control Policies

NIA Guidance Concerning Research on Human Participants

NIA Guidance on Sharing Data and other Resources

NIH is in the process of updating its general data-sharing policy from 2003. In the interim, it has developed separate policies on Genome Sharing and on Model Organism Sharing. NIA has also separately developed a policy on Alzheimer’s disease genomic sharing that is related to the broader NIH genome-sharing policy. For research related to these areas investigators are advised to follow the appropriate policy.

For all other research, NIA has developed the data-sharing guidance below.

Currently the 2003 NIH policy requires an agreed data-sharing/resource-sharing plan for awards of $500k and over. It allows flexibility below the 500k threshold for individual FOAs. Many of NIA’s currently open FOAs require data-sharing or resource-sharing plans regardless of the requested budget. In the future almost all FOAs will require a sharing plan.

The guidance below lists a uniform set of five elements to be addressed in the sharing plan. These elements are a starting point. Investigators may go beyond them as they develop plans. The points described below do not limit a plan.

Data: As defined here, “data” are final data on which a publication is based, usually presented in summary form in a publication These may include image libraries, video and audio recordings, or other derivatives of data that are relevant to the research being reported. In general, sufficient materials should be made available to allow replication and/or productive secondary analysis of the data. It is expected that raw data or minimally processed data would be shared.

  1. Investigators should indicate the repository/consortium/enclave where they will place the data or other resource or have a clear plan and capacity to share from their own site(s). Working with an external repository or through a consortium can be advantageous for many investigators. Others may have reasons for managing sharing at their own institution. Investigators should describe any transformations to the data (application of algorithms, tools) that will be applied to the data in preparing them to be shared.
  2. Investigators should describe and justify the timetable to release the data or other resource. Generally, the data and referenced resource from a publication should be made available by the on-line publication date unless NIH policy specifies an earlier date. Investigators should justify exceptions to that timing.
  3. Investigators should identify any restrictions on sharing (beyond temporal restrictions) and justify them. Restrictions might include, for example, no commercial use of the data or resource, only qualified users may access the material, or, no attempt to reveal personal or private information may be made. Investigators may also set different levels of access, for example, for expert users and novice users.
  4. For studies involving human participants, the investigators should describe how they will deidentify the data, whether that procedure is one following HIPAA, IRB, or other Institutional guidance, and generally how they will secure the personal and private information of the participants.
  5. Investigators should describe how they will curate the data or other resource and what documentation they will provide (“metadata”), in what format, and whether any data or resource will be held back or, should agree to work with the identified repository/consortium/enclave to prepare the data or resource and documentation, to allow easy access to qualified users.

FAIR (Findable, Accessible, Interoperable, and Re-usable) data principles

A large consensus panel of mainly European based authors derived the FAIR principles and explained them in this foundational article. Among the points they make beyond establishing these guiding principles are (1) That data should be findable by machines as well as by humans, as machines are far more efficient than humans at aggregating data sources, and (2) A trend towards special purpose repositories may be detrimental to true sharing especially if the repositories are not interoperable.

The steps outlined in elements 1 to 5 above are consistent with the FAIR principles. Yet they do not themselves ensure sharing that is as universal as the principles outlined in the 2016 article. We encourage investigators to explore these principles and we welcome sharing plans that do fully comply with these important principles.

Discussion points

Consent Forms (human data)

Institutional Review Board approved consent forms need to contain language that will allow the broadest possible sharing of deidentified data to investigators unaffiliated with the parent study. The language should avoid undue restrictions on sharing and be consistent with the data-sharing plan prepared for NIA.

Protecting privacy (human data)

Element 4 above does discuss the need to de-identify data when making them available for broad sharing. Some of the challenges to deidentifying now apparent include the software-aided ability to reconstruct faces from brain images, or the possibility of re-identifying individuals by combining research data with administrative data. NIA recommends putting a condition on sharing where the user certifies that no attempt will be made to reidentify participants from deidentified data.

Longitudinal studies

Longitudinal studies present additional challenges in deidentification as a consistent record of participant identity must be maintained through multiple waves of data collection. These problems usually can be resolved with an appropriate coding procedure. Issues concerning participants recruited using prior versions of consent forms that did not allow broad sharing should be discussed with program staff. In some circumstances it may be appropriate to reconsent participants.

Costs of making data or other resources available for sharing

Investigators should include funds for sharing as part of the competing application budget. Curation costs, deidentification costs, reconsenting costs, and other costs of sharing can be substantial.

Common data elements, combining data

As part of the data-sharing plan investigators should discuss whether the results might be part of a larger dataset eventually and what steps they might take to aggregate and harmonize the data to facilitate that. Such steps might include contacting a data repository prior to beginning the research to identify common data/study elements or methods that the investigator may use to allow combining data with other sets when there would be a clear advantage to combining data. It is a question that is most germane for those investigators who will host their data-sharing effort on-site. NIH, and other sites, also maintain resources of common data elements used in some research areas.

Review of Progress

NIA staff will consider progress in sharing as part of the review of progress during an award. Lack of progress in sharing is a reason to consider delaying approval of the next year of support. For renewal applications that request $500,000 or more of direct costs in a single year prior progress in sharing will be a factor in considering whether to accept the application.

HHS is committed to making its websites and documents accessible to the widest possible audience, including individuals with disabilities. We are in the process of retroactively making some documents accessible. If you need assistance accessing an accessible version of this document, please reach out to the guidance@hhs.gov.

DISCLAIMER: The contents of this database lack the force and effect of law, except as authorized by law (including Medicare Advantage Rate Announcements and Advance Notices) or as specifically incorporated into a contract. The Department may not cite, use, or rely on any guidance that is not posted on the guidance repository, except to establish historical facts.