NIH’s New Data Sharing Policy: Not So Scary
NIH now requires a formal data sharing plan for each grant they receive. (A few exceptions, primarily T32s and F grants, exist.) It sounds complicated, but they’ve actually made it pretty easy. You’re probably already doing some of what’s required.
First of all, there’s a template with reasonable descriptions of what to put in each piece. This is a great place to start. NIH has also created 12 example plans for different types of data such as human genomic data, clinical data, tech development, and more. (Most are 2-3 pages, so not too scary.)
“Scientific data” must be shared. This includes everything needed for someone to replicate your experiments or findings. Some journals now request this information, so providing it may be familiar. But before the thought of submitting everything related to your research overwhelms you, note that NIH has excluded anything not necessary to validate your findings, including, among other things, lab notebooks, preliminary analyses, and plans for future research. They’ve also defined fairly broad justifiable reasons to limit sharing of data. Examples of data types to be shared.
If you’re new to data sharing, you may be wondering how one puts data out there for the rest of the scientific world to use. NIH has a guide to selecting a data repository as well as a long list sorted by NIH institute of their preferred repositories.
The process for publishing a paper now includes sharing your scientific data “no later than the date on which the article is first made available,” which for most journals means when it first comes out online. If you have data that hasn’t been published but has been documented (e.g., conference proceedings, preprints), you must share it by the end of your grant performance period. If you have null results that don’t have a hope of being published, this is your chance to share them.
You’ll also need to identify which entity at your institution will monitor your compliance with your plan. Typically this is your grants office.
NIH has an extensive FAQ page that delves into the nitty-gritty of what, when, where, and how to both share your data and write your data sharing plan. There’s also a specific team to contact with further questions.
Data sharing is the first ingredient for replication, which can only help science progress.