top of page
Doctor_s Appointment

Data Led Prioritisation - reducing clinical risk and managing capacity

Data-led prioritisation is a methodology that enables effective prioritisation of patients on a clinical list through the automated and periodic application of clinical rulesets to routinely held hospital data.

The Challenge

Since the beginning of the COVID-19 pandemic, NHS waiting lists for new appointments have grown from 4.24 to 6.63 million*, caused by the redistribution of resources to acute care settings, staff sickness and reduced availability of inpatient beds**.

Similarly, outpatient clinic wait times within the UK have increased significantly following the COVID-19 epidemic*, and often the challenge of meeting follow-up demand on existing patients with chronic conditions (e.g. diabetes) reduces the capacity for services to accept new patient referrals.

In response to these mounting problems, NHS England have produced guidance that outlines a strategy to reduce wait times for elective care with a key goal being to prioritise vulnerable patients within waiting lists***.

However, within many trusts the only way to identify high risk and/or vulnerable patients is through manual review of patient notes, which is unfeasible given the constraints on clinical resource; the size of the lists; and the continued demand on services. Furthermore, should these patients be identified the capacity to treat them more urgently is often not readily available.

A data-driven system to identify vulnerable patients would therefore be of significant benefit; minimising risk and optimising resource utilisation by enabling the clinical prioritisation of patients who are likely to deteriorate whilst awaiting an appointment; whilst simultaneously identifying lower risk patients who might be suitable for lower intensity pathways (e.g. PIFU).

* Ref: BMA advice and support

** Ref: News

*** Ref: and NHS England

The Value Case

The effective prioritisation of patients can lead to a compelling value case;

  • reducing clinical risk

  • reducing unwanted variation

  • creating and freeing up capacity

  • reducing waiting times, particularly for the highest risk

  • starting to tackle health inequalities.

Our Methodology

When clinicians review patients’ records, they can (and often do) find information indicating that a patient needs to be seen urgently, even though they may not be scheduled to be seen for some time. In other cases, they find information that clearly shows that a patient is doing fine even though they may be scheduled to be seen soon. However, given the considerable pressures on their time, it is highly unlikely that they will be able to review all of their patients’ records in this way, even as a one-off exercise (let alone on a regular basis). 

Our methodology is to work with clinicians to capture some of the basic “rules” that they are using when reviewing records, and to apply these rules reliably at scale.  The results are fed back to the service so the information is available to support practical patient management (and service planning).

The rules can be very simple and intuitive e.g. “Since I last saw this patient, have they (possibly unknown to me) attended A&E with a complaint related to the condition I’m working with them on?” Or "Have more blood tests been taken that I should review?" 


Once defined, they can be applied quickly across the whole list to identify patients who are displaying some high risk markers and who might benefit from being seen sooner; or patients whose data suggests they are low risk and can be seen later (or even moved onto an alternative pathway such as PIFU).

A further advantage of this data-driven approach is that these rules can be applied to all data within the trust at a given point in time and can detect events (e.g. a related A&E admission) that clinicians were not previously aware of because they relate to events post-dating the last consultation with the patient. The system therefore acts as an early warning mechanism for any new and concerning data.

Our Solutions

The Waitlist Prioritisation Tool applies clinical intent to routinely held Trust data in order to provide a prioritised view of your entire waitlist. The tool:

  • Applies expert Clinical Intent drawn from a library of criteria developed by clinical specialists across the UK (or the intent can be authored by the consuming specialty if so desired). Clinical intent is intuitively designed to mirror clinicians review and risk assessment processes; this is not a “black box”.

  • Drives off standard data items held within hospital systems (appointments; lab tests; A&E visits etc.), that may not have been available to clinicians at the last point of contact with patient, ensuring latest up-to-date view on the patient’s condition. Data can also be augmented with primary care data where this is also accessible.

The output of the prioritisation can be consumed in a number of ways:

  • As a data product, with patient risk segments (and any supporting data) provided back to Trust data environments for integration with existing processes and/or systems

  • As a clinical review tool, which enables clinicians to quickly review the relevant clinical data in one place to aid in the prioritisation decision

  • As an administrative support tool which enables service admin staff to more efficiently action clinicians prioritisation recommendations by defining relevant pathways and assigning patients to them.

Case Studies

Please go to the Case Studies page for more information on where our approach has been successful.



This approach has first been used in Diabetes Outpatients at Guy's and St Thomas' NHS Foundation trust.

A number of papers have been published and presentations delivered on this emerging approach;

A pragmatic health informatics systems approach for aiding clinical prioritisation in a hospital based cohort of 4013 people with diabetes

J. Karalliedde*, O. French**, A. Smith**, L. Newcombe**, B. Malhotra*, C. Spellman*, N. Patel*, P. Sen Gupta*, D. Kariyawasam*, D. Rajasingam*, S. Thomas*

* Guy's and St Thomas' NHS Foundation Trust, London, UK; ** Factor 50, Nottingham, UK.

Oral presentation given at European Association for the Study of Diabetes Meeting 2022, Stockholm

Information on the presentation

Link to the Abstract

This work was also presented at the Association of British Clinical Diabetologists - link here

A pragmatic digital health informatics based approach for aiding clinical prioritisation and reducing backlog of care: A study in cohort of 4022 people with diabetes

J . Karalliedde*, O. French**, A. Smith**, L. Newcombe**, G. Burnhill**, C. Spellman*, M. Jessel*, D. Rajasingam*, S. Thomas*, A. Ayotunde*, B Malhotra*

* Guy's and St Thomas' NHS Foundation Trust, London, UK; ** Factor 50, Nottingham, UK.

Published July 2023 in Diabetes Research and Clinical Practice

PDF and Link

bottom of page