Predicting social care costs: a feasibility study

BARDSLEY Martin, et al
Nuffield Trust
Publication year:
Place of publication:

This report describes a study that explored whether statistical models can be used to predict an individual person’s future need for intensive social care in the UK. The aim of the project was to obtain pseudonymous individual-level data from several primary care, secondary care and social care organisations; link collate and analyse these data at the individual level; and attempt to develop a statistical model to predict which individuals are at greatest risk of requiring intensive social care in the 12 months after prediction. Data was provided by 4 Primary Care Trusts (PCTs) and one care trust. The research shows how it is possible to link routine data from health and social care information systems in a way that protects individuals’ identities. The project showed that it is possible to construct predictive models for social care. How these models might fit into everyday working practice now needs investigating. The predictive accuracy of the models was comparable to some of the models used by the NHS to predict hospital admissions. The authors comment that linked person-level information has the potential to improve quality of care services, whether through improved identification of high-risk individuals, comparative performance measures, service evaluations or budget-setting. There is a need to ensure that the quality of information about social care services improves comparably to the recent improvement seen in the quality of data about individual health care use.

Subject terms:
models, older people, planning, risk, service users, social care, social care provision, statistical methods, health care;
Content type:
United Kingdom
Register/Log in to view this resource

Key to icons

  • Free resource Free resource
  • Journal article Journal article
  • Book Book
  • Digital media Digital media
  • Journal Journal

Give us your feedback

Social Care Online continues to be developed in response to user feedback.

Contact us with your comments and for any problems using the website.

Sign up/login for more

Register/login to access resource links, advanced search and email alerts