The combined predictive model uses accident and emergency, inpatient, outpatient and GP data sources to identify future frequent users of hospital services beforehand, allowing for earlier intervention. The National Health Service (NHS) says that, with the right intervention, the deterioration of a patient’s condition could be prevented or slowed down.
As the system uses a combination of primary and secondary care data sources, the model can also categorize people with long-term conditions according to their risk of hospital admission, enabling organizations to implement different interventions and care pathways to meet these needs.
The King’s Fund and its partners, Health Dialog and New York University, were commissioned by the UK Department of Health and the strategic health authorities in 2005 to develop a number of techniques to accurately predict the future frequent users of hospital services. The first tool, the patients at risk of re-hospitalization (PARR) case-finding tool, was launched last year. According to the King’s Fund, the combined model builds on learning from PARR but draws on a much larger data source.
King’s Fund chief executive Niall Dickson commented, Previous techniques only allowed us to identify patients who had already been admitted to hospital on at least one occasion. However, this model allows us to go beyond this group to identify and provide better care for the vast numbers of people whose conditions are not yet at this critical stage. This more sophisticated approach will be crucial not only to provide better care but also to make better use of NHS resources.
