Skip to main content
Clinical Effectiveness Group

Compass (Discovery Data Service)

Compass is a database of deidentified primary care data, curated by CEG. It holds an extract of North East London data from the Discovery Data Service.

CEG is approved by NHS North East London Integrated Care Board (ICB) as a data processor, and holds an extract of primary care data from the Discovery Data Service in a deidentified database – Compass. 

We use Compass data for research projects with appropriate approvals from the ICB and the Joint Research Management Office (JRMO) for Barts Health NHS Trust and Queen Mary University of London. Through Compass, we have supported a wide range of research relevant to primary care, population health and reducing inequalities, using the information recorded in health records as part of routine care. 

Collaborating with CEG on research projects using Compass

If you would like to use Compass data for a research project, this must be done in collaboration with our team. To pursue this, please complete an Initial Project Idea form to tell us about your proposed research. The form will be reviewed by our internal Research Data Management Group. If your initial idea is approved by us, you will then work with your CEG co-investigator to complete a Compass Data Access Form, seek any outstanding approvals, and take the project forward.

All research projects using Compass need sufficient funding to support operating costs, data extraction, data wrangling, staff time, etc. This might be via funding already secured by your research team, or via future funding applications in collaboration with us.

We are not able to take forward all requests. If we decline a proposal, this may be due to capacity, or because your study does not align with our team’s current research priorities. In either case, we will provide feedback if we are not able to proceed.

Please contact the team if you have any queries: wiph-ceg-rdmg@qmul.ac.uk 

About the data

Compass holds high-quality, coded data covering demographics, diagnostics, symptoms, investigations, process of care, primary care consultations and prescriptions issued. It covers the whole NHS North East London region, which comprises seven 'Places': Barking and Dagenham, City and Hackney, Havering, Newham, Redbridge, Tower Hamlets and Waltham Forest. 

Compass has:

  • Full data coverage across all GP practices in NHS North East London from 2017 onwards
  • Deidentified data for 2.5 million currently-registered patients
  • Deidentified historic data for c.4.5 million patients who have left the area or died from 2017 onwards
  • SNOMED coded data, plus mapped Read2, CTV3, and local system codes
  • Pseudonymised patient NHS Number and UPRN address identifier, to enable data linkage
  • Daily data refreshes from GP clinical systems

Compass does not have:

  • Patient data or registration histories for GP practices outside the North East region
  • Data on patient attendance at A&E, hospital stays or outpatient clinics
  • Free text clinical notes or non-coded information
  • Synced, or Real-Time, data with GP practices – expected data latency is 24-48 hours

CEG can support linkage of Compass to external databases with the necessary ethics approvals in place.

Note: The processes of transferring and compiling data across different systems can produce variation in reported counts and measures. Data reports obtained from Compass should not be combined or directly compared with reports from GP clinical systems.

Which types of studies are suitable?

The Compass dataset can be used for research that:

  • Is across multiple practices, up to the entire NHS North East London region
  • Requires complex cohort building or querying
  • Is based on data from a specified date or timeframe
  • Requires non-aggregated extracts for use in analysis
  • Requires linkage to another dataset using NHS Number or Unique Property Reference Number (UPRN)
  • Cannot be better addressed by using QResearch or CPRD

Which studies are not suitable?

The Compass dataset cannot be used for research that:

  • Requires direct contact with patients or practices
  • Is designed to monitor or assess GP practice performance, either individually or grouped such as in PCNs or Federations
  • Requires identifiable patient information

Eligibility criteria

To collaborate with CEG on a research project using Compass data:

  • Your research must be publishable in a peer-reviewed journal
  • You must have funding to cover the cost of your study, or be applying for funding
  • The project team will need to include: 

1. A Co-Investigator from CEG
2. Someone who is medically qualified and has a clinical understanding of the subject
3. A statistician who contributes to the design of the study and will advise on the analysis
4. Someone with experience in electronic health record data research and knowledge of relevant clinical coding
5. At least two core members with a substantive post at Queen Mary

  • An individual team member may fulfill more than one of these roles
  • Approval for use of Compass data must be obtained from: CEG Research Data Management Group, Queen Mary University of London (via IRAS and JRMO) and North East London ICB.

Ethics and Information Governance

CEG has a unique and trusted relationship with GP practices across North East London, built over 25 years of collaborative working. We have agreement from all NHS North East London practices (the data controllers) to use their data for service improvement and research and development.

We provide data for research with the intention to benefit patients at individual, group or population level. We do not provide data for commercial purposes, for example insurance or drug marketing.

Where researchers have a cohort with the necessary consents in place, we have processes that enable linkage using pseudonymised NHS numbers and UPRNs.

We consider that certain conditions – for example; HIV, sexually transmitted diseases, abortions, teenage pregnancy and use of Lower Layer Super Output Area Codes – are particularly sensitive and require consideration of governance and agreement before any such data is extracted.

We expect that third party users will:

  • Comply with the Five Safes Framework; completing the relevant NHS Data Security and Protection Toolkit and other relevant training and complying with these standards.
  • Where appropriate, be inducted in use of the data extract and/or clinical systems by a member of CEG staff. 

Acknowledgement

All published research using Compass needs to acknowledge the data source with the following statement:

‘This work uses data provided by patients and collected by the NHS as part of their care and support, and with the support of general practitioners across North East London. The data were made available with approval of data controllers by the Clinical Effectiveness Group, Queen Mary University of London, who provided a curated deidentified dataset for research purposes.’

 

Back to top