Use the Cohort Discovery Tool to find out how many Stanford patients match a particular clinical phenotype, and learn more about this set of patients. Only aggregated, approximate results are provided so this can be used in the preparatory to research phase. For example, you might determine whether there are enough patients to justify a particular study.
Many search criteria are available for defining cohorts, including structured data (diagnosis and procedure codes, drug classes and ingredients, etc.), unstructured clinical notes, and temporal (time-based) constraints. Counts and visualizations are available for each cohort you define, and you can compare multiple cohorts.
Saving and Sharing Searches
Once you have created a search you wish to keep, click on "Searches" in the upper left and select "Save Search". If you wish you can share with a collaborator by ticking the "Save to public library" checkbox. To rename a previously saved search, use "Save Search As...".
To find a cohort definition saved to the public library, click the "Searches" button and select "Public Library".
The Help menu option in the upper right has several useful links:
- "Read the documentation" takes you to the home page of this site
- "Request a consultation" takes you to the Data Science Resources Consultation intake form in ServiceNow
- "View data timeliness report" shows how up-to-date the underlying clinical data is
- "Tell us what you think" opens up a feedback form
You must be on the Stanford physical network or Virtual Private Network (VPN) to access this tool, and you must have an appropriate role within the School of Medicine.
Furthermore, as of September 2021, you are required to have both a fully sponsored SUNetID and Cardinal Key to access STARR Tools, as the login process now requires you to first authenticate to Google using your firstname.lastname@example.org identity. We are aware of the hardship these new requirements are causing and are working to remediate.