Choosing a predictive risk model: a guide for commissioners in England

Predictive risk models are used for predicting events such as unplanned hospital admissions, which are undesirable, costly and potentially preventable. Such models have been shown to be superior to other ‘case finding’ approaches, including threshold models and clinical opinion. Although the Department of Health has previously funded two predictive models for the NHS in England, the current policy is to promote an open market in terms of suppliers of risk tools. Commissioners should consider a range of factors when choosing whether to ‘make or buy’ a predictive model, including the outcome to be predicted, the accuracy of the predictions made, the cost of the model and its software, and the availability of the data on which the model is run. Predictive models should be seen as one component of a wider strategy for managing patients with chronic illness. Although there are opportunities here for improving the health status of patients with complex needs while making net savings for the NHS, the evidence for hospital-avoidance interventions is patchy and therefore robust evaluations should be built into any proposed local strategies. In the future, it is unclear whether predictive risk models in England should best be procured or built at a local, regional or national level.

Choosing a predictive risk model: a guide for commissioners in England | The Nuffield Trust [Internet]. [cited 2013 Apr 24]. Available at: http://www.nuffieldtrust.org.uk/publications/choosing-predictive-risk-model-guide-commissioners-england

None