Readmission Risk Prediction

Increasing the knowledge of acute hospital capacity management.

Whilst much health policy focus has largely been on services within hospital, the issue of how to improve care for people with long term medical conditions has been rapidly climbing up the policy agenda. Improving the management of high cost patients, especially those with long term conditions, is increasingly seen as an important strategy for improving health outcomes and controlling healthcare expenditure. The World Health Organisation (WHO) describes care for long term conditions as “the health care challenge of this century”, with such conditions currently responsible for 60 per cent of the global burden of disease and likely to be the leading causes of disability by 2020. Research has shown that a minority of “frequent flyer” patients account for a disproportionately large proportion of health expenditure. This situation may become more pronounced as people are living longer with increasingly complex conditions and as the actual number of people with a long term condition increases. This work is about putting in place ‘upstream’ care to prevent the deterioration of individuals’ conditions to the point where an expensive acute emergency admission is required. This research would involve data analysis identifying high volume/high risk patients, targeted services and informed patient-level clinical intervention, reviewing workforce requirements, and implementing evidence-based strategies for improved productivity.

Key Research Questions

  1. Can the patient’s risk of readmission be accurately predicted?
  2. Can the risk prediction be completed in near real-time (ideally before discharge planning commences)?
  3. Can a care team put appropriate interventions in place based on tool output?
  4. Can readmission rate be reduced as a result of ongoing risk assessment and care planning?
  5. Can the prediction model be improved based on data collected as part of ongoing risk assessment and care planning?