HUB-API - Clinical Deterioration Prediction Modelling

  • Research type

    Research Study

  • Full title

    Development and internal validation of risk prediction models to identify impending clinical deterioration in Acute Inpatient and Emergency Department populations, using machine learning and artificial intelligence methods.

  • IRAS ID

    344974

  • Contact name

    Sasidharan Sameer

  • Contact email

    research@clini-hub.co.uk

  • Sponsor organisation

    Northumbria Healthcare NHS Foundation Trust (NHCT) Research and Development Team

  • Clinicaltrials.gov Identifier

    NIO0002, Sponsors Number

  • Duration of Study in the UK

    1 years, 0 months, 0 days

  • Research summary

    In hospitals, the National Early Warning Score (NEWS-2) is a health alarm for patients. Nurses and doctors regularly measure observations including heart rate, blood pressure and oxygen levels. These measurements are used to calculate a score which has been shown to indicate the likelihood of a person becoming more unwell and requiring a review by a doctor. NEWS-2 is used currently in the UK. It is intended to help detect problems early, however the current research evidence suggests that there might be some limitations: It relies on vital signs from a single snapshot in time and may miss other important signs of deterioration (e.g. subtle changes in observations over time). NEWS-2 has been developed to predict whether a patient may be sicker in 24 hours’ time. Whilst NEWS-2 is a valuable tool for identifying deteriorating patients the recognised limitations may be mitigated by the development of more advanced models that leverage machine learning or deep learning algorithms. Clini-Hub Limited seeks to develop one or more models that work to support the identification of deteriorating patients in a hospital environment with better statistical performance than NEWS-2. It is recognised in research literature that these algorithms may be better at spotting when a patient's health is getting worse, as they are designed to find complex patterns in data. This research forms part of a wider series of projects that contribute towards the development of software to support clinical teams in identifying patients who are becoming sicker and need medical review. Clini-Hub Limited, an independent data analytics company, are engaging with an NHS Trust for this specific research project to ensure that the developed models are scientifically rigorous and clinically applicable. Clini-Hub is responsible for funding and conducting the research.

  • REC name

    North East - Tyne & Wear South Research Ethics Committee

  • REC reference

    24/NE/0122

  • Date of REC Opinion

    20 Aug 2024

  • REC opinion

    Further Information Favourable Opinion