KePPOD
Research type
Research Database
IRAS ID
227612
Contact name
Chris Farmer
Contact email
Research summary
Kent Predicted Patient Outcomes Database (KePPOD)
REC name
South Central - Oxford C Research Ethics Committee
REC reference
18/SC/0158
Date of REC Opinion
8 Jun 2018
REC opinion
Further Information Favourable Opinion
Data collection arrangements
Data collection for KePPOD will include data from East Kent Hospitals University NHS Foundation Trust (EKHUFT) all electronic Health Record clinical information systems covering all acute hospital admissions and outpatient attendances over the previous five year period (Calendar Years 2013-2017 inclusive).
Data will be collected directly by members of the EKHUFT IT team who have routine access to all data concerned. Only retrospective, routine data will be collected; no data will be collected from patients solely for the purpose of inclusion in KePPOD. Data will be de-identified by the EKHUFT IT team (which may include subcontractors) prior to transfer to UoK for storage of the dataset. Project-specific re-hashing of pseudonymous identifiers will also be undertaken in UoK for each project using the dataset, ensuring that applicants only receive fully de-identified data from KePPOD.
Key files linking the dataset to personal information will not leave EKHUFT and will remain in the control of the Trust’s Caldicott Guardian, this will be held to allow for any incidental clinical findings to be acted upon or data removed from KePPOD if patients wish to withdraw their data.
Research programme
The KePPOD database will support high quality research into the use of healthcare data to predict risk and patient outcomes to better inform clinical decision making. The database was set up to support three initial projects. These projects are summarised below and will be completed within the initial 5-year ethics approval period. 1. Testing machine learning algorithms predicting patient care events in order to develop, evaluate and implement systems and processes to reduce harm events in the acute hospital setting. 2. Improve patient care delivered in the acute setting by providing insight into patients’ clinical and social care needs using a novel approach of combined hospital datasets and mobile application technology. 3. To develop a tool for women with pre-existing chronic kidney disease (CKD) which estimates the amount of kidney damage likely to occur during pregnancy. PREDICT (PREgnancy-associated progression of chronic kidney Disease: development of a Clinical predictive Tool). It is intended that the database will be made available to researchers to maximise the research outputs of creating this dataset. The research programme is not limited to the use of data for risk or outcome predication. Use of the dataset will be overseen by a Database Committee.
Research database title
Kent Predicted Patient Outcomes Database (KePPOD)
Establishment organisation
University of Kent
Establishment organisation address
Centre for Health Services Studies
University of Kent
Canterbury
CT2 7NZ