Risk modelling in the critically ill Version 1.0
Research type
Research Study
Full title
Risk modelling for quality improvement in the critically ill: making best use of routinely available data
IRAS ID
172505
Contact name
David A Harrison
Contact email
Clinicaltrials.gov Identifier
NCT02454257, ClinicalTrials.gov registration
Duration of Study in the UK
2 years, 0 months, 0 days
Research summary
Summary of Results
Risk prediction models use information from early in a patient’s illness to predict their likely outcome. They are useful for benchmarking clinical services and for research. The Intensive Care National Audit & Research Centre (ICNARC) is a registered charity that benchmarks services for adult critical care (intensive care)—the ICNARC Case Mix Programme—and patients that experience a cardiac arrest during their hospital stay—the National Cardiac Arrest Audit.
In her Annual Report of the Chief Medical Officer 2011, Professor Dame Sally Davies highlighted the potential “to do much more, particularly through the linkage of existing data”. In this study, we will link together existing datasets in order to improve the risk prediction models used for benchmarking critical care and cardiac arrest services.
Current risk models predict whether a patient will survive to be discharged from hospital. However, this is far from the end of the journey for patients that have been critically ill. Data linkage to death registrations from the Office for National Statistics will enable us to extend our models to predict longer term mortality. Also, many patients that survive intensive care experience ongoing poor health. Data linkage with the National Diabetes Audit and UK Renal Registry will enable us to better understand which patients go on to develop diabetes and renal (kidney) failure. For patients undergoing cardiac (heart) surgery, data linkage with the National Adult Cardiac Surgery Audit will provide additional information from before and during the operation to improve risk predictions for patients admitted to a cardiothoracic critical care unit and explore how risk changes along the patient journey. Finally, linkage with routine Hospital Episode Statistics will enable us to estimate the cost of subsequent hospital care as well as to determine the contribution of chronic health conditions to outcomes for patients experiencing a cardiac arrest.
Summary of Results
: Large amounts of information (data) are collected about patients using NHS services, but we do not make the best possible use of these data to improve patient care. Data are held by different organisations in different databases. Joining up these databases (data linkage) can give us a more complete picture of what happened to a patient.
The Intensive Care National Audit & Research Centre is an independent charity that runs national clinical audits to monitor and improve care for critically ill patients. These audits use statistical models that take information about the patient know before, or soon after, the start of their illness to make a prediction of their likely outcome. In this research study, we used data linkage to improve these models and ensure that the audits provide useful information back to hospitals to support quality improvement. However, it took over 4 years to link the databases.
By linking with death certificate information, we were able to predict how many patients die by 30 days, 90 days and 1 year after their critical illness. By linking with routine hospital data, we were able to take better account of how sick patients were before they became critically ill and look at how many days they spent in hospital in the year after their critical illness and the costs of these hospital stays. By linking with two other national clinical audits, we were able to develop new models to predict important problems of kidney failure and diabetes that some patients experience after critical care. By linking with another national clinical audit, we were able to get a more complete picture of how sick patients having heart surgery were before they were admitted to an intensive care unit, helping us to improve our models to make fairer comparisons for these patients.
REC name
Wales REC 5
REC reference
15/WA/0256
Date of REC Opinion
17 Jul 2015
REC opinion
Favourable Opinion