Writing an NIH Data Management & Sharing Plan

Decisions to make when sharing research data

The decisions about what data can be shared, with whom, how/where, and when are complex and inter-related. Depending on applicable regulatory requirements, federal, state, and local laws, you may have to address multiple requirements.  The NIH DMS Policy expects that you will maximize appropriate sharing of the data needed to validate and replicate your findings.

The first step is to identify all relevant legal and ethical obligations and seek appropriate input from experts in that area. The Considerations for Sharing Research Data at IU form is designed to assist with this.  The NIH sharing site can help you determine if additional NIH ICO (Institute, Center, or Office) policies apply using the Which Policies Apply to My Research? tool.

Once you have identified the constraints or fixed decisions about data sharing, you can identify what options are available and make the remaining decisions.

DMS Plan Section Instructions

Give plans and timelines for data preservation and access, including:

• The name of the repository(ies) where scientific data and metadata arising from the project will be archived. See Selecting a Data Repository for information on selecting an appropriate repository. 
• How the scientific data will be findable and identifiable, i.e., via a persistent unique identifier or other standard indexing tools.
• When the scientific data will be made available to other users and for how long. Identify any differences in timelines for different subsets of scientific data to be shared.
  • Note that NIH encourages scientific data to be shared as soon as possible, and no later than the time of an associated publication or end of the performance period, whichever comes first. NIH also encourages researchers to make scientific data available for as long as they anticipate it being useful for the larger research community, institutions, and/or the broader public.

For data subject to the Genomic Data Sharing (GDS) Policy

• For human genomic data:
  • Investigators are expected to submit data to a repository acceptable under the Genomic Data Sharing Policy. See Where to Submit Genomic Data.
  • Human genomic data is expected to be shared according to NIH’s Data Submission and Release Expectations, but no later than the end of the performance period, whichever comes first. 
• For Non-human genomic data:
  • Investigators may submit data to any widely used repository.
    Non-human genomic data is expected to be shared as soon as possible, but no later than the time of an associated publication, or end of the performance period, whichever is first.

The NIH DMS policy requires that the scientific data necessary to replicate your findings be shared when possible in an established data repository to make the data FAIR (Findable, Accessible, Interoperable, Reusable). When writing a DMS plan, it will be necessary to think about what repository will be the home for the data that you anticipate creating. Some data will be best preserved in an NIH-affiliated repository, and some will not. Some data may need to be de-identified before sharing is appropriate (please see Privacy & Confidentiality for de-identification considerations).

The guidance given by the NIH describes the following decision path:

NIH ICO Requirements

First, check to see if the NIH Institute, Center, or Office (ICO) funding the award asks for a specific repository.  Check the Funding Opportunity Announcement to see if a specific repository is required as part of the award itself. For example, the National Institute of Neurological Disorders and Stroke (NINDS) provides a list of NINDS-Supported Data Repositories.  Please see this list of NIH ICO data sharing requirements for ICO-specific policies.

Discipline-Specific Repositories

If there is a repository which houses data related to the discipline of the research, plan to share in that kind of repository when possible.  Check also on whether a discipline-specific repository exhibits evidence of the NIH's desired characteristics listed in the box below this one. 

• NIH-supported Scientific Data Repositories This section of the NIH data sharing website can be used to locate discipline-specific repositories which are also funded whole or in part by the NIH.  Other discipline-specific options may be available depending on your area of research.
• re3data The Registry of Research Data Repositories categorizes data repositories by subject and content type. 

Generalist Repositories

An established generalist repository, containing data related to many disciplines, or an institutional repository can then be selected.  Several popular generalist repositories are part of the NIH-led Generalist Repository Ecosystem Initiative (GREI).  Of these, we recommend first looking at Figshare and Dryad for general data storage needs, and Vivli for clinical research data.  We name these platforms here to advocate for those that we think have the best chance of persistence over time.  For generalist repository feature comparisons, please see this Generalist Repository Comparison Chart.

Institutional Repositories

Our institutional data repositories meet the desirable characteristics of repositories articulated by the NIH and provide public or open access to data.  In most cases, there is no charge for depositing and preserving data for faculty, staff, and students at IU. Neither institutional repository accepts datasets containing protected data elements, such as Personal Health Information (PHI) or Personally Identifiable Information (PII).

• IUPUI DataWorks - A data curation expert will help you prepare your data for sharing, including writing a README file, creating metadata, and more. 
• IU DataCORE A pilot project repository offered by the IU Libraries for all types of research data in any file format. 

Choosing a data repository that fits well with DMS Policy guidance involves considering the terms of the award and the research domain represented in the data.  A specific repository may be required by a funding NIH ICO (Institute, Center, or Office) policy or FOA (Funding Opportunity Announcement).  If no specific repository is required, a discipline-specific repository, specializing in a particular research domain, is encouraged in the interest of promoting discoverability and data reuse.  If no discipline-specific repository exists, a generalist (Figshare, Dryad, Vivli) or institutional repository (DataWorks) may be selected.

Any repository should be considered on the basis of how/if it exhibits the following desirable characteristics:

• Unique Persistent Identifiers
  • Is there a DOI (Digital Object Identifier) generated by the repository for each dataset or other asset?
  • Is there a persistent link generated that directly connects to the dataset once it has been published, or is there no direct URL?

• Long-Term Sustainability
  • Does the repository's policy explain how funding will be managed sustainably?
  • Is there a description of contingency plans to preserve the infrastructure of the repository?

• Metadata
  • Are there descriptions and examples of the schema used to describe the data the repository hosts?
  • Are any standards referenced, and do these have a basis in best practices for the scientific discipline(s) represented by the repository's contents?

• Curation and Quality Assurance
  • What elements are recommended or required as part of a dataset submission?
  • Is there a review process before a dataset is accepted, and if so, what is it?

• Free and Easy Access
  • Is there a timeline listed for when open datasets can expect to become available after submission?
  • What evidence exists that the repository is responsive to controlled access data requests?

• Broad and Measured Reuse
  • How does the repository describe licensing options for datasets?
  • Is all information needed to appropriately attribute and generate citations for datasets present in the public view of the repository site?

• Clear Use Guidance
  • Are terms and conditions for reuse addressed in submission documentation and/or the public view of hosted datasets?
  • Is there documentation describing how a data reuse request is handled?

• Security and Integrity
  • How is identity verified when requests are made to access controlled datasets?
  • Once a dataset is live on a repository's servers, are regular checksums and backups performed?

• Confidentiality
  • For sensitive data, what capabilities does the repository have to maintain confidentiality?
  • If the repository asks for de-identified data, how is de-identification defined and monitored?

• Common Format
  • Does the repository prefer open source data and file formats over proprietary formats where possible?
  • Is there evidence that the formats the repository supports are those most commonly used within the discipline represented by the scientific data?

• Provenance
  • Does the repository's architecture protect a dataset from being altered or deleted after being accepted?
  • If accepted dataset alteration is possible, is the original dataset still preserved by the repository, or is there any other description of version control methods?  

• Retention Policy
  • Are datasets ever removed from a repository after any period of time?
  • If a dataset owner wants to remove a dataset, is this possible?

IU Libraries provide support for researchers seeking NIH-funding. We are actively monitoring updates and guidance available from the NIH, as well as the IU Research Data Management Plan Working Group. We will continue to update the information here to reflect those changes. Our support includes:

• data management planning
• selecting the appropriate repository(ies) for sharing data
• planning for documentation that will enable data sharing, and
• some aspects of preparing data for deposit
• support decision-making related to data management and sharing that occurs during project start-up and active conduct of research

We will also connect you with other experts across IU who can help you submit the best DMS Plan possible. We are not able to write DMS Plans on behalf of research teams of which we are not members.

To request a consultation, contact the data librarian for your campus or school:

• Heather Coates, IUPUI - hcoates@iu.edu 
• Levi Dolan, Indiana University School of Medicine - dolanl@iu.edu
• Ethan Fridmanski, Indiana University Bloomington - ejfridma@iu.edu