Over the years, the NIH has released policies to encourage data sharing. Our first data sharing policy in 2003 was limited to awards requesting more than $500,000 in direct costs in any year. Subsequent data sharing policies focused predominantly on genomic data sharing (Genome-Wide Association Studies Policy; Genomic Data Sharing Policy). There are clear and well-accepted advantages of data sharing, including increasing sample size, facilitating reproducibility analyses, and increasing the impact of the taxpayer’s investment in data collection to advance science. These advantages are not limited to large grants or to genomic studies; they are applicable to data sharing of nearly all of the research that the NIH supports. Therefore, after considering initial input from the scientific community, the NIH this week released its Draft NIH Policy for Data Management and Sharing that is intended to encourage data sharing of all NIH-supported research.

The draft policy proposal attempts to address the many challenges to data sharing. Not all data are the same and the NIH Institutes and Centers have some flexibility in the types of mechanisms they can use to support data sharing. The draft policy acknowledges that there can be legal, ethical, and privacy constraints on data sharing, and these constraints can be incorporated into the data management and sharing plan. The plan is to be submitted with Just-In-Time materials so that only those grant applications to be funded are required to develop a plan, and the draft policy includes supplemental draft guidance on the Elements of a NIH Data Management and Sharing Plan. Appropriately sharing data also involves costs to store, annotate, and make the data available to other researchers; therefore, the draft policy includes supplemental draft guidance on Allowable Costs for Data Management and Sharing so that these costs can be incorporated in the budget.

The Associate Director for Science Policy, Carrie Wolinetz, published a blog earlier this week on this draft policy, encouraging the scientific community and other interested stakeholders to review the NIH’s current proposal and provide comments. So why am I also writing a blog on the same topic for the behavioral and social sciences research community? The NIH is predominately a biomedical research organization, and when considering policy implementation and its impacts, there is a tendency to consider predominately biomedical research examples. The behavioral and social sciences often collect highly sensitive information (e.g., substance abuse and mental health diagnoses and treatments, child maltreatment, HIV status) that can be difficult to deidentify (e.g., video recordings of parent-child interactions), and that may be constrained by the data source (e.g., Tribal agreements, state vital statistics agreements). The NIH has worked hard to consider as many variations as possible of data types and sources and how these fit within this draft data management and sharing policy, but I am sure we have not been exhaustive. Therefore, I encourage every behavioral and social science researcher to review this draft policy, consider how a future policy will affect the research you conduct and how you store and share data, and provide detailed comments on this draft policy.

You can provide comments here. The deadline for submitting comments is January 10, 2020. Thanks in advance for taking the time to review this draft policy and make comments on it as it pertains to the research you conduct. It is only with your input that the NIH can finalize a data management and sharing policy that has considered all of the various ways that NIH-supported research is conducted and that addresses all of the variations of data management and sharing necessary to implement this policy in a way that advances our commitment to responsible data sharing.