Skip to main content

Secure analytics platform for NHS electronic health records

OpenSAFELY delivers research across over 58 million people's health records, always respecting patient confidentiality

Working in partnership with

  • University of Oxford logo
  • Nuffield Department of Primary Care Health Sciences logo
  • London School of Hygiene & Tropical Medicine logo
  • TPP logo
  • EMIS logo

Better research, improved patient confidentiality

Why OpenSAFELY?

OpenSAFELY is a highly secure, transparent, open-source software platform for analysis of electronic health records data. All platform activity is publicly logged. All software for data management and analysis is shared, automatically and openly, for scientific review and efficient re-use.

Protecting privacy

We never let researchers download patient data, and OpenSAFELY tools let users to write code to analyse patient data without even needing to view the raw records.

Core privacy features

Auditable by the public

It is a privilege to use patient data for the public good. OpenSAFELY respects patients by carefully considering this in every part of its design.

How do I know my data is safe?

Better, open science

OpenSAFELY requires publication of all analytic code, and our tools drive all users to produce prespecified, reusable, testable, shareable and modular software for research.

View best practices

Enabling high volumes of research

Over 60 published research outputs to date, with many more in progress.

Research outputs

OpenSAFELY is revolutionising the way we work with health data. The initiative has already delivered major contributions to public health during the COVID-19 pandemic. But OpenSAFELY also leads the way in terms of transparency, open working methods, and a uniquely secure data access model.

Krishnan Bhaskaran

Professor of Statistical Epidemiology, London School of Hygiene and Tropical Medicine

Ready to learn more? Get started within an hour

Open for research

We are currently working with NHS England to cautiously on-board a small number of external pilot users to develop their analyses on OpenSAFELY. This first wave of pilot users will be collaborators, working closely alongside us to co-develop the platform.

Read about our pilot onboarding process
  1. Python, R and Stata

    Choose any of these languages to write your analytic code

  2. Primary care data

    Our deployment for NHS England uses primary care data to research questions related to the Covid-19 emergency

  3. A commitment to open ways of working

    We believe transparency and open working methods are key to earning public trust and improving research quality

Latest from the blog

  1. How OpenSAFELY began

    A short history of OpenSAFELY’s beginnings, as published in our booklet, “The past, present and future of OpenSAFELY”

  2. Earning and maintaining trust: PPIE and more

    How we engage with patients and the public, as published in our booklet, “The past, present and future of OpenSAFELY”

  3. The legal basis: ethics, controls and building trust

    A brief review of information governance, as published in our booklet, “The past, present and future of OpenSAFELY”

  4. Output checking helps to keep private data safe

    A brief look at our output checking service how it works, and why it’s necessary. As published in our booklet, “The past, present and future of OpenSAFELY”

  5. Standard tools for data preparation, and federated analytics

    A brief look at how we enable federated analytics, as published in our booklet, “The past, present and future of OpenSAFELY”

  6. Co-pilots give newcomers a helping hand

    A brief overview of our co-piloting programme, as published in our booklet, “The past, present and future of OpenSAFELY”

View more blog posts →