Streamlining Data Sharing: Strengthening Your DMS Plans with GREI's Practical Guide

This presentation is part of a three-part webinar series hosted by the Generalist Repository Ecosystem Initiative (GREI), designed to assist scientific researchers in navigating data management and sharing requirements The series provides an introduction to GREI-developed resources that streamline the data-sharing lifecycle, from planning to submission, helping researchers maximize the impact of their work.

In this specific session, "Strengthening Your DMS Plans with GREI’s Practical Guide" , presenters Julie Goldman (Harvard Library/Harvard Dataverse) and Ana Van Gulick (Digital Science/Figshare) introduce the Guide for Including a Generalist Repository in an NIH Data Management and Sharing Plan. The slides walk through the core elements of the NIH DMS policy and provide practical strategies, including sample text, for incorporating generalist repositories into Elements 1 through 5 of a compliant plan.

The slides also provide practical user stories that demonstrate real-world applications of data sharing, discovery, and reuse. Specifically, the presentation features case studies from Renee Parks (Washington University in St. Louis) on sharing obesity prevention policy data to advance health equity; Jim Brorson (University of Chicago) on harmonizing stroke recurrence data across multiple clinical trials; and the Research Coordinating Center for Health and Extreme Weather (CAFÉ RCC) team (Boston University School of Public Health & Harvard T.H. Chan School of Public Health) on building and supporting a data-sharing community for climate and health research. Key insights include:

• Planning for data sharing during the initial study design phase is critical, especially ensuring that participant consent materials clearly explain how data might be shared
• Open science and participant privacy can balanced by sharing non-sensitive supplementary materials openly while keeping the primary, sensitive datasets under restricted, controlled access 
• Harmonizing anonymized data from multiple sources can achieve the statistical power necessary to build new models and detect novel patterns.
• Centralized, well-documented data collections help overcome common hurdles—such as finding specific spatial or temporal data—which significantly lowers the barrier to entry for early-career researchers
• Providing ready-to-use analysis code, software, and robust documentation alongside datasets ensures reproducibility and allows others to confidently explore the data from new angles