Information governance

Dr Foster Unit at Imperial College was set up in 2002 and is part of Imperial College London. Our core funding comes from the NIHR Imperial Biomedical Research Centre (BRC). We are a small unit consisting of researchers, PhD students, 2 Data managers and a Project Administrator. The unit is led by Paul Aylin  and Alex Bottle- Professors of Epidemiology and Public Health.

We are a research unit and almost all of our research is based on information held in hospital administrative records about the care you received in hospital. We do not have access to your clinical notes. We use your data to produce and analyse measures for investigating variations in quality of care that patients receive.

The Dr Foster Unit at Imperial College London holds records about the health care and treatment you have received in any English NHS hospital in the form of administrative data (Hospital Episodes Statistics) – inpatient and day case admissions, outpatient appointments and Accident and Emergency attendances. These records are linked to electronic death certification information (ONS Mortality files, which we also hold separately) that includes the date and causes of death.

These data are purely administrative data containing summary information on diagnoses and operations and do not include the full clinical notes. We do not receive or hold any identifiable data such as names, date of birth or addresses in any of our files at any time.

The Dr Foster Unit’s mission is to help NHS organisations improve the standards of care that patients receive. We do this in several ways. The first is by analysing the information described above to highlight variations in the quality of care. For example, we have shown that patients admitted as emergencies at the weekend have lower survival rates than those admitted for the same diseases during the week; We also found that survival following planned surgery is lower when the operation is done on a Friday than when it is done on a Monday, potentially due to lower staffing levels at the weekends. These findings have been used in the debate about how to make the NHS work more 24/7.

The second way we help the NHS drive up standards is by using your hospital information in a system for spotting when hospitals have higher than expected mortality rates. This system flagged Mid Staffordshire hospital trust as having problems in 2007 and led to the national regulator inspecting that trust, amid a local campaign by relatives, leading to a public inquiry into poor standards there. We alert hospital chief executives when their trusts have high mortality rates for a range of diagnoses and procedures. Our alerts system forms part of the Care Quality Commission’s hospital inspection regime.

The third way we help improve care is to use the data for researching the best ways to measure the quality of care and to compare hospitals and treatments. Mortality is important, but there are many other important things to look at such as readmissions, infections and other complications of treatment. We put a lot of emphasis on patient safety. We work on collaborative projects with clinical colleagues to help us develop and validate healthcare quality indicators other than mortality, including measures for bariatric (weight loss) surgery, primary angioplasty rates, stroke care, obstetric care, orthopaedic redo operation rates and unexpected returns to the operating theatre. We work with hospitals to help them understand this information in their efforts to improve standards.

We believe in transparency and publish our statistical and other methods in peer-reviewed journals and on our website. Our academic output from this research is high profile and has already helped to drive forward healthcare policy. As an example our research on weekend hospital death rates has helped to change services to provide better care at weekends.

We understand that patient information can be sensitive and we can assure you that only the right people have access to your patient information. Protecting your personal confidential information is central to all that we do and we treat your data with great care and respect.

All health data we hold are held on a secure computer system, with no physical links to the internet. Only a few named researchers are allowed access to the data, and they have all signed strict confidentiality agreements. We are fully compliant with national standards that permit us to hold patient level data.

We do not share your data with any third party and we take care that any of the outputs or analyses that we publish cannot disclose the identity of an individual.

All our records are retained as stipulated in our contracts. We will not keep your records for longer than necessary. All records will be destroyed confidentially once their retention period has been met, and when the unit has made the decision that the records are no longer required. 

We hold no identifying information about you so are unable to identify you from our datasets. However, you can choose whether or not your data is used for research and planning. Details of how to opt out of sharing your data are available at this link here: https://www.nhs.uk/using-the-nhs/about-the-nhs/opt-out-of-sharing-your-health-records/

If you have a concern about the way your records are managed or to learn more about how we use, manage and maintain confidentiality of your information, please contact:

Richard Cooley
Database Manager
Dr Foster Unit
School of Public Health
Imperial College London
Reynolds Building
St Dunstan’s Road
London W6 8RP

Our full policy can be found in the Dr Foster Unit GDPR Privacy Notice (PDF)