Cosmos Frequently Asked Questions

You can request access to Cosmos here.

As of May 22, 2026, you must first complete the Cosmos Researcher badge before completing the Cosmos Super User badge.

We strongly recommend that all Cosmos users complete the Cosmos Researcher and Super User badges (Train Tracks). These trainings are free, can be done at any time, and should normally require only a couple of hours to complete. The Super User Badge provides a solid introduction to the system and opens up 10 more Cosmos seats.

For researchers who haven't yet done Cosmos training and are attempting to start with the Researcher badge, you will see the "Oops! We Couldn't Find Your Training Record--Let's Fix That!" error message after clicking "Self-Assessment" → "Take the Assessment Now." To fix this, please email training@epic.com and request that a training profile be created with the Cosmos Researcher curriculum attached.

There is no cost to use Cosmos to follow hunches, ask questions, track trends, and more. Fees from sponsored projects fund the continued growth and maintenance of Cosmos. For a summary of the Cosmos training fees that are required for line-level access, see "How do I gain line-level access to Cosmos data?" below.

If using Cosmos data as part of a sponsored project, we advise that you engage the Epic Cosmos team during the budgeting process to discuss any applicable Cosmos Study Fees. We recommend creating a project in the Cosmos portal and disclosing that it’s funded in response to whatever questions Cosmos asks. After that, someone from Epic will follow up with you on how to proceed and any potential costs.

Depending on the size of the sponsored project,  fees may not apply.

First, you will need to request access to Cosmos. You must have Cosmos access in place before registering for Cosmos training.

For Data Science Virtual Machine (DSVM) access, you must then complete the Cosmos Data Model for Data Architects or Data Scientist certificate:

• Aggregate-level access:
  Use Slicer Dicer in the Cosmos Portal for cohort discovery and feasibility analysis.
  
  
• Data Science Virtual Machine (DSVM) access:
  Line-level access (SQL):
  Requires Epic COS305 (0.5 day) + COS500 (1.5 days) (estimated $800 total*) (Data Model for Data Architects certificate).
     
  Line-level data science tools access (R and/or Python):
  Requires Epic COS305 (0.5 day) + COS550 (0.5 day) (estimated $400 total*) (Data Scientist certificate).
  
  *Researchers are responsible for funding their own Cosmos training. Epic will accept a PTA or Cost Center number as a payment method.
  
  Courses must be taken sequentially, and all advanced Cosmos training can be completed virtually.
  
  

After completing the Cosmos Data Model for Data Architects or Data Scientist training, email cosmosteam@stanford.edu with proof of training completion. We will then check that you have the correct SHC Epic profile assigned. Once confirmed, we will ask Epic to turn on your access to the Cosmos DSVM (EDDI).

You'll find a more detailed explanation of Cosmos access levels and required training in Epic Galaxy. Epic oversees all Cosmos training and certification. Stanford is only the intermediary.

No line-level data may ever be exported from Cosmos: Working with the Data: Data Science Virtual Machine Cosmos will only let researchers transfer aggregate data files or images after manual review by Epic: Publish or Share Your Findings

Epic has a process for importing an .xls / .csv file (and other file types) into Cosmos. Please contact us at cosmosteam@stanford.edu for more details.

The data in Cosmos is dynamic in nature, so old encounters can still be updated as things such as chart corrections or additional documentation happen after the fact. The longer it has been since the visit, the less likely such an update will occur, but there is still a possibility.

If an encounter is updated on the source system after being sent to Cosmos, it will be re-queued for sending to Cosmos and would appear as though it is “refreshed” the next time the Cosmos dataset is updated

If you do not log in to Cosmos for over six months (180 days), you will be blocked for inactivity. If this happens, please contact us at cosmosteam@stanford.edu to request that your Cosmos account be reactivated.

If you do not log in to Cosmos for over a year, we will deactivate your account and make the seat available to new Cosmos users.

You can avoid this issue by logging in to Cosmos at least once every six months.

Your Cosmos account will be automatically deactivated after you leave Stanford.

We recommend that you review the Cosmos Publication Checklist. The use of Cosmos data in a publication requires citation as part of the user agreement.

Cosmos is only for Stanford-related research.

You can not use Cosmos for non-Stanford-related work. You must confirm that you agree that your use of Cosmos will only be for Stanford-related work before obtaining access.