Skip to content

Introduction

OpenSAFELY is designed to allow you to do your analytic work on your own computer, without ever having to access the real, sensitive, patient-level data.

Because the tutorial study uses dummy patient data, anyone can complete the tutorial.

We ask all potential collaborators to complete this tutorial before applying to run their project against real data.

Learning outcome🔗

After completing this tutorial, you will have seen the steps needed to write an OpenSAFELY-compliant study that runs with "dummy" (randomly-generated) patient data.

You will:

  1. Create a GitHub account
  2. Create a new GitHub repository
  3. Create a study that queries "dummy" patient data to generate a dataset
  4. Run the study in a development environment
  5. Update the study to add extra data columns to the dataset
  6. Add an analysis step to the study
  7. Run the study in a testing environment

Please ask any questions in our Q&A forum!

Requirements🔗

The requirements for this tutorial are minimal.

You will need a computer with:

  • an up-to-date web browser
  • an internet connection

The tutorial will be completed in the GitHub Codespaces environment. Free GitHub accounts are given a monthly quota of Codespaces. If you do not have a GitHub account, we will create one in this tutorial.