Using REDCap for NIH Data Sharing

REDCap & NIH Data Archive (NDA) Guidance

If your study includes data sharing or archiving requirements with the NIH Data Archive (NDA), it is critical to plan your REDCap project accordingly from the start.

NDA submissions require structured, standardized, and well-documented data. Designing your REDCap project with these requirements in mind will significantly reduce rework and ensure successful data submission.

Important: If your study requires NDA submission, you cannot treat REDCap as a standalone data collection tool. Your project must be structured to support downstream data sharing, validation, and formatting requirements.
Core Principle: Design your REDCap project with the end data submission format in mind.

What is the NIH Data Archive (NDA)?

The NIH Data Archive (NDA) is a centralized repository for sharing and preserving research data, particularly for studies funded by the National Institutes of Health (NIH).

NDA requires:

  • Standardized data structures
  • Defined data elements
  • Consistent coding and formatting
  • Proper handling of identifiers and subject-level data

Why This Matters in REDCap

REDCap projects that are not designed with NDA requirements in mind often require significant restructuring before submission.

  • Variable names may not align with NDA data elements
  • Response options may not match required coding
  • Data may need to be reshaped or reformatted
  • Identifiers may not be handled appropriately
Best practice: Align your REDCap data dictionary with NDA data structures before building.

Key Design Considerations

  • Use standardized variable naming conventions
  • Match response choices to NDA-required coding when applicable
  • Avoid unnecessary free text fields
  • Ensure consistent data formats (dates, numeric values, etc.)
  • Document field definitions clearly
Best practice: Build your REDCap project as if it were already in its final export format.

Handling Identifiers & PHI

NDA submissions often involve subject-level data that must be carefully managed and de-identified according to NIH requirements.

  • Do not use direct identifiers as record IDs
  • Clearly flag identifier fields in REDCap
  • Separate identifiable and non-identifiable data where appropriate
  • Follow IRB and institutional policies for data sharing
Important: REDCap does not automatically prepare data for NDA submission. Proper handling of identifiers is the responsibility of the study team.

Data Structure & Standardization

  • Ensure consistent variable formats across all records
  • Use coded values instead of free text where possible
  • Understand whether NDA expects wide vs long format
  • Plan for repeat measures and longitudinal data structure
Best practice: Standardization at the time of data collection is far more efficient than post hoc data cleaning.

Testing & Validation

  • Export sample data early in development
  • Compare exports to NDA requirements
  • Validate coding, formats, and structure
  • Identify gaps before data collection begins
Best practice: Perform a “mock NDA submission” using test data before go-live.

Resources & Guidance

Tip: If your project includes an NIH Data Management & Sharing Plan (DMSP), ensure your REDCap build aligns with those requirements from the outset.