The whole IORD database is stored on a big NHS computer. Only a very small number of people, just 3 to 5, can get into it. These people all work for Oxford University, and the NHS has given them special permission to work on the whole database.
No one else gets to see everything in IORD. But other researchers can see parts of it if they follow specific rules. You can find out more about how this works in IORD application trajectory.
Before a researcher starts looking at any data, they need to work out what they want to learn. Imagine you’re going on a treasure hunt, but you need to know what kind of treasure you’re looking for before you start. Here are three types of questions they might ask:
- Descriptive questions: These are asking, “Tell me about this.” For example, we might want to know if the number of people getting measles has been going up in recent years. We’re trying to describe what’s been happening.
- Predictive questions: These questions are all about guessing what might happen in the future. It’s like trying to predict the weather. For example, can we figure out which patients are likely to get very sick and need intensive care in a hospital?
- Explanatory questions: These questions are trying to understand why something happens. For instance, does having COVID-19 antibodies in your body protect you from getting COVID-19 again? We’re trying to find the reasons behind what is happening.
Before a researcher starts a project, they need to decide if they want to learn about what’s been happening, what might happen, or why something happens. This helps them focus on what they are looking for and what information they need. So they start by writing a plan explaining what they want to do. The plan needs to say exactly what specific information they need, and who they need this information from, to answer their question.
You can find a list of all the questions researchers have tried to answer in the IORD Projects section.
These plans are checked by a team including people from the hospitals and GPs who don’t have anything to do with IORD, and members of the public. This group of experts and everyday people think about:
- Who is the researcher, and are they for real? We want to make sure they’re from a trustworthy organization and have a history of doing medical research.
- Is the question they’re asking important and connected to infections? We don’t want to share data for something that doesn’t really matter.
- Does their plan make sense? We need to be sure the way they want to answer their question is right.
- Are the people they want information from, and the specific pieces of information they’re asking for, related to the question they’re trying to answer? Every person’s data and every piece of information they ask for should be used in their research.
If the group agrees, then the researcher only gets the data they need to answer their questions.
- For instance, let’s say there is a project about infections in people with a certain kind of cancer. In that case, we would only give the researcher the information from people in IORD who have that exact type of cancer. And they’d only get the particular information they need to answer their questions, nothing extra. It’s like giving them just the puzzle pieces they need for their specific puzzle.
- Other projects look at certain types of infections, like bloodstream infection caused by a microbe called Escherichia coli. We would only share the information from people in IORD who had Escherichia coli in their blood with the researcher. They’d only get the details they need to find their answers, like giving them the right pieces to solve their puzzle.
- Once in a while, a project needs information from a lot of people in the database, maybe even most of them. For instance, if we’re trying to work out why some people are more likely to get an Escherichia coli bloodstream infection, we need details from both the people who got this infection and those who didn’t. This way, we can figure out what things might make someone more at risk. It’s like doing a big puzzle where we need lots of pieces to see the whole picture.
Only non-business groups can submit plans, like charities, non-governmental organizations, and universities, but they can come from anywhere in the world. IORD does not sell data to anyone – at the moment we are able to provide data for research related to infections from non-business groups without charging.
Information for researchers
- If you are a researcher interested in using IORD data, you can contact iord@ndm.ox.ac.uk to get a proposal template.
- To access the data, you’ll need to have your proposal reviewed and approved by the Research Database Team.
- You’ll also have to agree to keep the information confidential and follow certain rules about how it’s used.
- If you’re not working directly for the NHS, you might need a special agreement with your organization to make sure you handle the data responsibly.
- After careful consideration by the IORD research database team and the IORD patient and public panel, the consensus is that requests from students (at any level) that are primarily for educational purposes, including reproducing and enhancing skills, are not considered in scope for IORD and will not be approved. This is because the underpinning legal basis for the ethical approval for IORD, including section 251 support for the use of data without individual consent via the Confidentiality Advisory Group, is “task in the public interest”. Using data for personal learning is not directly in the public interest and therefore does not meet GDPR requirements for the legal basis for processing the data without consent. Personal learning can be achieved through other data and therefore use of IORD data for this purpose does not meet the Caldicott principles. Of note, in addition a Data Sharing Agreement would be required with an institution who would take legal responsibility for use of the data in a way that safeguards patient confidentiality. Reproducing results is in the public interest, but this should be the stated aim of an institutionally supported research project.