HSMA previous projects — full collated corpus

Generated: 2026-08-10 by previous_projects/collate_projects_for_llm.py
Source: every index.qmd under previous_projects/hsma_*/ in the HSMA website repository (https://github.com/hsma-programme/hsma_site)
Projects in this file: 102
Cohorts: HSMA 1 (6), HSMA 2 (10), HSMA 3 (6), HSMA 4 (16), HSMA 5 (13), HSMA 6 (49), HSMA Alumni (2)

About this document

HSMA (Health Service Modelling Associates) is a training programme in which health, social care and policing staff learn simulation, machine learning and operational research techniques and apply them to a real problem in their own organisation. Each entry below is one such project.

How this document is structured

Each project is a level-2 heading of the form ## <Project ID> — <Title>, and projects are separated by a horizontal rule.
Project IDs encode the cohort and project number, e.g. H6-6015 is project 6015 from HSMA 6. They are stable and can be cited.
Under each heading is a bulleted metadata block, then the project’s Abstract (a short summary written for the website listing), then the Project description (the full page text).
Abstract and description text is reproduced verbatim from the source pages, including any British spellings and Quarto shortcodes such as . Only the metadata has been reformatted.
Techniques and Application areas come from a controlled vocabulary used across the site, so they are reliable for filtering and grouping.
Projects the site keeps in an ARCHIVE folder are excluded from this file, so it reflects the main project listing.
An index of every project follows; the full entries begin after it.

Project index

Project ID	Cohort	Title	Techniques	Application areas
H1-1000	HSMA 1	Modelling A & E flows to inform redesign at North Devon District Hospital	Discrete Event Simulation (DES)	Emergency Departments, Patient Flow
H1-1001	HSMA 1	Modelling the resources needed to run an acute frailty unit	Discrete Event Simulation (DES)	Frailty, Acute care, NHS 10-year plan shifts: Hospital to Community
H1-1002	HSMA 1	Modelling resources needed to reduce out of area placements in the Mental Health Acute Care pathway	Discrete Event Simulation (DES)	Mental Health, Patient Pathways, NHS 10-year plan shifts: Hospital to Community
H1-1003	HSMA 1	Modelling the impact of various strategies to improve weekend discharge rates	Discrete Event Simulation (DES)	Discharge, NHS 10-year plan shifts: Hospital to Community
H1-1004	HSMA 1	Modelling the impact of regionalising cardiac arrest centres	Geographic Modelling	Cardiology, Discharge
H1-1005	HSMA 1	Modelling the potential impact of having a Clinical Decision Unit	Discrete Event Simulation (DES)	Emergency Departments, Reducing Backlogs, NHS 10-year plan shifts: Hospital to Community
H2-2001	HSMA 2	Using modelling to understand the delays in mental health services in Devon	Discrete Event Simulation (DES)	Mental Health, NHS 10-year plan shifts: Hospital to Community
H2-2002	HSMA 2	RCHT Eldercare workforce mapping to identify needs and alternatives to cover the frailty pathway	Discrete Event Simulation (DES)	Frailty, NHS 10-year plan shifts: Hospital to Community
H2-2003	HSMA 2	Artificial Ian – can a machine learning tool learn when it is necessary to cancel surgery? And can we use geographic modelling methods to improve outpatient clinic utlilisation in the community?	Machine Learning	NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Hospital to Community, Surgery
H2-2004	HSMA 2	Reducing the delays to glaucoma treatment at Torbay Hospital	Discrete Event Simulation (DES)	NHS 10-year plan shifts: Sickness to Prevention, Opthalmology
H2-2005	HSMA 2	Using modelling to optimise catheter lab efficiency and predict stroke bed demand	Discrete Event Simulation (DES)	Hospitals, Demand & Capacity
H2-2006	HSMA 2	Improving ambulance response times for life-threatening emergencies using simulation modelling	Discrete Event Simulation (DES)	Ambulance
H2-2007	HSMA 2	Can ambulance dispatch codes be used to determine when an ambulance is really needed?	Discrete Event Simulation (DES)	Ambulance, NHS 10-year big bet: AI to drive productivity
H2-2008	HSMA 2	Reducing waiting times for spinal patients using simulation modelling	Machine Learning	Reducing Backlogs
H2-2009	HSMA 2	Testing interventions to reduce the Continuing Healthcare Assessment backlog	System Dynamics	Reducing Backlogs
H2-2010	HSMA 2	Developing a model to understand delays to discharge	Discrete Event Simulation (DES)	Discharge, Hospitals, NHS 10-year plan shifts: Sickness to Prevention
H3-3000	HSMA 3	What are they saying about us? An AI tool to determine the sentiment of tweets to police forces across the country, and what people are talking about	Sentiment Analysis, Natural Language Processing (NLP)	Police
H3-3001	HSMA 3	Generating a richer understanding of relationships in crime data in order to identify opportunities to safeguard individuals and families	Network Analysis	Police, Crime
H3-3002	HSMA 3	Modelling strategies to reduce the elective backlog in hip surgery	Discrete Event Simulation (DES)	Elective Surgery, Orthopaedics
H3-3003	HSMA 3	Developing a generic vaccination service model for the COVID-19 pandemic and beyond	Discrete Event Simulation (DES)	Vaccination, COVID-19, NHS 10-year plan shifts: Sickness to Prevention
H3-3004	HSMA 3	Exploring the use of Machine Learning and Natural Language Processing to teach a machine to predict whether a patient is likely to be imminently admitted to hospital based on GP data and clues in GP notes	Natural Language Processing (NLP), Machine Learning	Inpatient Admissions, Primary Care (GP), Hospitals, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to Deliver Impact
H3-3005	HSMA 3	Simulation modelling to test proposed models of pediatric critical care	Discrete Event Simulation (DES), Mapping, Location Optimization	Paediatric, Intensive Care Units & Intensive Therapy Units (ICU & ITU), Inpatients
H4-4000	HSMA 4	The Effect of Booked Appointments on Waiting Times at Urgent Treatment Centres	Discrete Event Simulation (DES)	Urgent Treatment Centres (UTC), Emergency Departments, ECDS Dataset, Booked Appointments in Walk-in Services, NHS 10-year plan shifts: Hospital to Community
H4-4001	HSMA 4	South East Regional Covid 19 Vaccination Demand & Capacity Modelling	Forecasting, Prophet, Plotly Dash	Vaccination, COVID-19, NHS 10-year plan shifts: Sickness to Prevention
H4-4002	HSMA 4	Use of Discrete Event Simulation to Tackle Long Waits and a Growing Backlog for Children Requiring Neuro Development Assessment (Autism and ADHD)	Discrete Event Simulation (DES)	Neurodiversity, Paediatric, Mental Health, Demand & Capacity, Reducing Backlogs, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
H4-4003	HSMA 4	Using DES to Improve Flow through an Acute Medicine Assessment Pathway	Discrete Event Simulation (DES)	Emergency Departments, Patient Flow, Inpatients
H4-4004	HSMA 4	Discrete Event Simulation of Cognitive Behavioural Therapy Pathway in an IAPT Service	Discrete Event Simulation (DES)	Mental Health, NHS Talking Therapies (Formerly IAPT), Community Mental Health
H4-4005	HSMA 4	Use Of Discrete Event Simulation (DES) to reduce delays in Cancer Diagnosis & Treatment	Discrete Event Simulation (DES)	Cancer, Colorectal, Identifying Bottlenecks in Pathways
H4-4006	HSMA 4	Discrete Event Simulation to Improve Flow and Performance in the Urgent Treatment Centre	Discrete Event Simulation (DES)	Urgent Treatment Centres (UTC), Emergency Departments, Identifying Bottlenecks in Pathways
H4-4007	HSMA 4	Forecasting Demand and Length of Stay in the Emergency Department	Machine Learning, Forecasting, Streamlit	Paediatric, Length of Stay, Emergency Departments, Inpatients, Staffing Level Optimisation
H4-4008	HSMA 4	Meeting the demand of 111 for primary care services	Discrete Event Simulation (DES), Plotly Dash, Network Analysis, Telephone-based Services	111 Service, Primary Care (GP), Emergency Departments, NHS 10-year plan shifts: Hospital to Community
H4-4009	HSMA 4	Using Machine Learning to Predict Hospital Admissions and Length of Stay for Respiratory Conditions	Machine Learning	Respiratory, Inpatient Admissions, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to deliver impact
H4-4010	HSMA 4	The role of Patient Initiated Follow-up (PIFU) and ‘Digital Outpatients’ in Supporting the Elective Recovery - Can We Better Size Potential for Clearing the Backlog?	Discrete Event Simulation (DES)	Patient Initiated Follow-Up (PIFU), Digital Outpatients, Rheumatology, Reducing Backlogs
H4-4011	HSMA 4	Predicting Non-Elective Admissions	Machine Learning, Natural Language Processing (NLP)	Non-elective Admissions, Inpatient Admissions, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to Deliver Impact
H4-4012	HSMA 4	Predicting Violent Incidents on Mental Health Inpatient Units	Machine Learning	Mental Health, Mental Health Inpatients
H4-4013	HSMA 4	Reducing Travel Times to Treatment for Cardiac Patients in the South East of England	Mapping, Travel Times, Streamlit, Location Optimisation	Cardiology
H4-4014	HSMA 4	Developing a Service Planning Decision Support Tool to Tackle Inequalities and Minimise Carbon Output	Machine Learning, Automation, Reproducible Analytical Pipelines (RAP)	Carbon Emissions, Health Equity Audits, Non-attendance Prediction, NHS 10-year plan shifts: Analogue to Digital, NHS 10-year big bet: Data to Deliver Impact
H4-4015	HSMA 4	Spatial Modelling of Violent Crime to Support Strategic Analysis	Mapping, Geostatistics, QGIS	Police
H5-5002	HSMA 5	Using Natural Language Processing to detect drug related content within free text	Natural Language Processing (NLP), Named Entity Recognition	Police
H5-5003	HSMA 5	Using Discrete Event Simulation to model the bottlenecks in the Acute Medical Unit pathway	Discrete Event Simulation (DES), Streamlit	Acute Medical Unit (AMU), Emergency Departments, Acute Care, Hospitals, NHS 10-year plan shifts: Hospital to Community
H5-5004	HSMA 5	Modelling the effect of complex discharge delays on acute performance	Discrete Event Simulation (DES)	Discharge, Hospitals, Patient Flow, Acute Care, Inter-service Interactions, NHS 10-year plan shifts: Hospital to Community
H5-5005	HSMA 5	Discrete Event Simulation to model elective surgery pathways	Discrete Event Simulation (DES), Streamlit	Elective Surgery, Waiting Times, Surgical, Hospitals
H5-5006	HSMA 5	Using Machine Learning to estimate inequities in access to hospital procedures	Machine Learning	Inequalities, Planned Admissions, APC Dataset, NHS 10-year plan shifts: Sickness to Prevention
H5-5008	HSMA 5	Modelling the location of neonatal critical care units in North West England	Mapping, Location Optimization, Discrete Event Simulation (DES)	Neonatal, Inpatients, Hospitals
H5-5009	HSMA 5	Network Analysis of diagnostic procedures in A&E setting	Network Analysis, Plotly Dash	Emergency Departments, Diagnostic Procedures, ECDS Dataset
H5-5010	HSMA 5	Creating a tool to automatically generate health equity audits for Community Diagnostic Centres	Streamlit, Automation, Reproducible Analytical Pipelines (RAP)	Health Equity Audits, Community Diagnostic Centres (CDCs), Inequalities, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Analogue to Digital
H5-5012	HSMA 5	Forecasting Demand – Investigating approaches to forecast clock starts	Forecasting, Prophet, ARIMA	Clock Starts, Demand & Capacity, Waiting Lists, NHS 10-year plan shifts: Analogue to Digital
H5-5013	HSMA 5	Understanding Excess Mortality in Dorset	Streamlit, Forecasting	PCMD Dataset, Public Health, NHS 10-year plan shifts: Sickness to Prevention
H5-5014	HSMA 5	Investigating factors impacting NHS workforce retention	Plotly Dash, Regression, Machine Learning	Staff Turnover, Workforce
H5-5015	HSMA 5	A Discrete Event Simulation Model to reduce Rheumatology waiting times in Dorset	Discrete Event Simulation (DES)	Rheumatology, Waiting Lists
H5-5016	HSMA 5	Developing a tool to assess inequalities and demographic coverage of service locations	Streamlit, Mapping, Travel Times	Inequalities, Demographics, NHS 10-year plan shifts: Sickness to Prevention
H6-6001	HSMA 6	Discrete Event Simulation modelling of Non-elective flow	Discrete Event Simulation (DES), Streamlit	Patient Flow, Non-elective Admissions, Emergency Departments, Same-day Emergency Care
H6-6002	HSMA 6	Geographic and Boosted Tree Modelling of Healthcare worker vaccination uptake	Geographic Modelling, Machine Learning	Vaccination, COVID-19, Workforce, NHS 10-year plan shifts: Sickness to Prevention
H6-6003	HSMA 6	DESmond: Discrete Event Simulation and Artificially Intelligent Forecasting: Modelling a Prostate Cancer Pathway	Discrete Event Simulation (DES), Streamlit	Cancer, Waiting Times, NHS 10-year big bet: Data to Deliver Impact, NHS 10-year plan shifts: Sickness to Prevention
H6-6005	HSMA 6	Improving ambulance care via fast feedback from Quality Care Indicators	Machine Learning, Natural Language Processing (NLP), Sentiment Analysis	Ambulance
H6-6006	HSMA 6	Discrete Event Simulation modelling of childrens ADHD diagnosis and treatment	Discrete Event Simulation (DES), Streamlit	Neurodiversity, Paediatric, Waiting Lists, Waiting Times, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
H6-6007	HSMA 6	Modelling eye injection pathways	Agent Based Simulation (ABS), Discrete Event Simulation (DES), Streamlit	Opthalmology
H6-6008	HSMA 6	Modelling GP Phone calls	Discrete Event Simulation (DES)	111 Service, Emergency Departments, Primary Care (GP), NHS 10-year plan shifts: Hospital to Community
H6-6009	HSMA 6	Using machine learning models to predict future frailty	Machine Learning	Frailty, Older Adults, NHS 10-year plan shifts: Sickness to Prevention
H6-6010	HSMA 6	Understanding drivers of increased length of stay	Causal Analysis, Machine Learning, Explainable AI, Synthetic Data, Streamlit, System Dynamics	Length of Stay, Inpatients, Understanding Drivers
H6-6011	HSMA 6	Forecasting modelling for A&E attendance	Forecasting, Reproducible Analytical Pipelines (RAP)	Emergency Departments, Demand & Capacity, Seasonality
H6-6012	HSMA 6	Using classification modelling technques to investigate changes in Healthcare Resources Group (HRG) coding over time	Machine Learning, Explainable AI	Clinical Coding, Costs
H6-6013	HSMA 6	Modelling bed occupancy on an Acute Ward	Forecasting, Machine Learning, Discrete Event Simulation (DES), Streamlit	Length of Stay, Inpatients, Bed Occupancy, Demand & Capacity
H6-6015	HSMA 6	Forecasting the supply of medical doctors	Forecasting	Workforce
H6-6016	HSMA 6	Optimising the location of Breast Cancer diagnostic services across Devon	Geographic Modelling, Travel Times	Cancer, Women’s Health
H6-6017	HSMA 6	Referral to treatment waiting times for Neurosurgical patients	Discrete Event Simulation (DES), Streamlit	Surgical, Neurology, Waiting Lists, Waiting Times
H6-6018	HSMA 6	Predicting the risk of injurious falls in older people with atrial fibrillation	Machine Learning, Explainable AI	Older Adults, NHS 10-year plan shifts: Sickness to Prevention
H6-6019	HSMA 6	Clinical coding automation using Natural Language Processing	Natural Language Processing (NLP)	Clinical Coding, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
H6-6020	HSMA 6	Modelling 111 option 2 call centre	Discrete Event Simulation (DES), Forecasting, Streamlit, Quarto	111 Service, Telephone-based Services, Staffing Level Optimisation, NHS 10-year plan shifts: Hospital to Community
H6-6021	HSMA 6	Predicting the future demand for Renal replacement therapy	Forecasting	Renal, Dialysis, NHS 10-year plan shifts: Hospital to Community
H6-6023	HSMA 6	Predicting Gestational Diabetes and other maternity-related conditions using machine learning	Machine Learning, Streamlit	Diabetes, Maternity, Women’s Health, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: AI to drive productivity
H6-6024	HSMA 6	Developing a streamlit app for creating Theographs of patient journeys	Data Visualisation	Patient Pathways
H6-6025	HSMA 6	Modelling delays in breast, head and neck cancer pathways	Discrete Event Simulation (DES)	Cancer, Women’s Health, Waiting Times
H6-6026	HSMA 6	Developing a DES Model for a Mental Health Hub	Discrete Event Simulation (DES)	Mental Health, Community Mental Health, Waiting Times, NHS 10-year plan shifts: Hospital to Community
H6-6034	HSMA 6	Forecasting blood donation session capacity	Machine Learning, Streamlit	Blood & Transplant, Non-attendance Prediction
H6-6035	HSMA 6	Developing a web app to recommend appropriate technology enabled care	Streamlit, Python	Older Adults, Costs, NHS 10-year plan shifts: Hospital to Community
H6-6036	HSMA 6	Forecasting demand in RDUH breast care services and the impact of urban development	Discrete Event Simulation (DES), Forecasting, Streamlit	Cancer, 2-week Wait, NHS 10-year plan shifts: Hospital to Community
H6-6037	HSMA 6	Mapping health inequalities, depreviation, ethnicities and crime across the UK	Geographic Modelling, Mapping	Inequalities, Crime, Police, Demographics, NHS 10-year plan shifts: Sickness to Prevention
H6-6039	HSMA 6	Applying Natural Language Processing to automate the extraction and classificiation of congenital anomaly diagnoses from free text and genetic data	Natural Language Processing (NLP), Streamlit	Congenital Anomilies, Genetic Data, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
H6-6040	HSMA 6	Modelling the benefit of MECC (Making Every Contact Count) Training using agent based simulation	Agent Based Simulation (MECC), Streamlit	Making Every Contact Count (MECC), Population Health, NHS 10-year plan shifts: Sickness to Prevention
H6-6041	HSMA 6	Identifying which patients are most at risk for an outcome across integrated neighbourhood teams	Machine Learning, Explainable AI, Streamlit, Geographic Modelling	Demographics, Public Health, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
H6-6043	HSMA 6	Predictive modelling for smoking cessation success	Machine Learning, Explainable AI, Causal Analysis	Smoking, Public Health, Council, NHS 10-year plan shifts: Sickness to Prevention
H6-6044	HSMA 6	Population segmentation of GP-registered population in Dorset	Machine Learning, Unsupervised Learning	Primary Care (GPs), NHS 10-year plan shifts: Sickness to Prevention
H6-6045	HSMA 6	Forecasting NHS planning and performance metrics	Forecasting	NHS
H6-6046	HSMA 6	Proactive Patient Attendance Prediction: Enhancing Healthcare Efficiency through Attendance Forecasting	Machine Learning	Non-attendance Prediction, Outpatients, NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Sickness to Prevention
H6-6047	HSMA 6	RALPulator : Predicting Robotic-assisted laparoscopic prostatectomy (RALP) operative times from patient letters	Machine Learning, Natural Language Processing (NLP), Streamlit	Urology, Cancer, Surgical, NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Sickness to Prevention
H6-6048	HSMA 6	Identifying potential concurrent treatment areas and services that would better support patients with multiple, complex referral to treatment (RTT) pathways.	Machine Learning, Streamlit	Inequalities, Patient Pathways
H6-6052	HSMA 6	Geographical mapping in specialist palliative and end of life care	Mapping, Geographic Modelling	End-of-life Care & Hospices, Demographics, NHS 10-year plan shifts: Hospital to Community
H6-6055	HSMA 6	DES Modelling of The Hyperacute / Acute Stroke Pathway - Patient and Economic Outcomes	Discrete Event Simulation (DES), Streamlit	Stroke, NHS 10-year plan shifts: Sickness to Prevention
H6-6056	HSMA 6	Redrawing North West Ambulance dispatch Boundaries	Geographic Modelling	Ambulance
H6-6058	HSMA 6	Building a Machine Learning tool to predict Did Not Attend (DNA) events	Machine Learning, Streamlit	NHS 10-year big bet: AI to drive productivity
H6-6060	HSMA 6	Optimising Same Day Emergency Care (SDEC) Resourcing	Discrete Event Simulation (DES), Emergency Departments	NHS 10-year plan shifts: Hospital to Community
H6-6061	HSMA 6	Analysis and forecasting of referrals into hospital	Forecasting	Neurology, Demand & Capacity, Referrals
H6-6062	HSMA 6	Developing a primary care load management tool	Discrete Event Simulation (DES)	Primary Care (GP), NHS 10-year plan shifts: Hospital to Community
H6-6063	HSMA 6	Modelling the Talking Therapies clinical pathway using a Discrete Event Simulation	Discrete Event Simulation (DES)	NHS Talking Therapies (Formerly IAPT), NHS 10-year plan shifts: Hospital to Community
H6-6065	HSMA 6	Automating injury coding using language models	Natural Language Processing (NLP), Machine Learning	Clinical Coding, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
H6-6066	HSMA 6	Evaluating the Impact of Community Diagnostic Centres on Health Inequalities and Patient Access to Diagnostic Services	Geographic Modelling, Streamlit, Python	Community Diagnostic Centres (CDCs), NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
H6-6067	HSMA 6	Applying and manipulating identification rules for specialised services	Streamlit, Python	Specialised Services, SUS Dataset
H6-6069	HSMA 6	Generating wordclouds from referral information	Natural Language Processing (NLP)	Referrals
H6-6070	HSMA 6	Using Machine Learning techniques to predict Hospital Readmissions	Machine Learning, Explainable AI	NHS 10-year plan shifts: Hospital to Community
H1-1100	HSMA Alumni	Geographic modelling of resource utlitisation at Devon Air Ambulance	Geographic Modelling, Forecasting, Discrete Event Simulation (DES)	Ambulance, Air Ambulance, Patient Transport, Waiting Times
H5-5100	HSMA Alumni	eFIT: Extra funding allocation - inequality tool	Streamlit	Inequalities, Costs

Projects

H1-1000 — Modelling A & E flows to inform redesign at North Devon District Hospital

Project ID: H1-1000
Cohort: HSMA 1
Authors: Nic Harrison (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Emergency Departments, Patient Flow
Public code repository: No
Source file: previous_projects/hsma_1/H1_1000_modelling_A&E_flows_to_inform_redesign/index.qmd

Abstract

The project aimed to enhance A&E performance by optimising staffing and bed requirements, focusing on the 4-hour target. Various staffing scenarios were modelled, revealing near-optimal staffing with minor adjustments. Success was achieved through collaborative efforts, meeting the 4-hour target in Q1-Q3 despite increased attendances. The project promoted system-wide improvements and cultural change within NDHT and Devon STP.

Project description

This project aimed to improve the performance of A&E services within the broader hospital system, considering factors such as the 4-hour target, staffing optimization, and bed requirements. The project focused on optimizing staffing within current time constraints to enhance A&E performance and the conversion rate from attendances to emergency admissions.

A base case was established, and various staffing scenarios were modelled, exploring approximately 40 possible configurations. The model was designed to be simple for ease of understanding while also complex enough to track patient types and key hospital processes. The main focus was on identifying pinch points in patient flow and improving 4-hour performance.

The analysis revealed that both nursing and doctor staffing were already close to optimal, but minor adjustments could improve performance. However, unsafe options or those that might affect recruitment and retention were eliminated. Optimizations were found within the current staffing budget, and these scenarios were tested during a “Perfect Fortnight” initiative, alongside 36 other improvement efforts.

The project achieved notable success, with the 4-hour target met in Q1, Q2, and Q3, despite a 6% increase in attendances and a national breach rate of 15%. In response to challenges in meeting the target for Q4, additional investments in doctor staffing were made, which, combined with other changes, helped to achieve the standard.

The project’s success was attributed to a collaborative approach, with the “embedded analysis” for A&E and the role of “modelling ambassadors.” The modelling work was shared with six other Trusts in the Devon STP and extended to other operational areas, promoting system-wide improvements. Cultural change within NDHT and Devon STP led to more integrated Business Intelligence (BI) efforts, influencing decision-making and operational practices across the region.

H1-1001 — Modelling the resources needed to run an acute frailty unit

Project ID: H1-1001
Cohort: HSMA 1
Authors: Joe Turner (Royal Cornwall Hospitals NHS Trust)
Organisations: Royal Cornwall Hospitals NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Frailty, Acute care, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_1/H1_1001_modelling _resources_to_run_acute_frailty_unit_Cornwall/index.qmd

Abstract

A discrete event simulation model to identify number of beds to improve care for frail older patients in the first 72 hours of emergency care. A model identified optimal bed numbers, showing an 18-bed unit would meet the 4-hour ED standard for 94% of patients, enhancing outcomes and reducing admissions.

Project description

As part of Royal Cornwall Hospitals Trust work with the Acute Frailty Network to “To improve the care of frail older people in their first 72 hours of emergency care, reducing admissions and number of bed days used whilst also improving the hospital experience.” It was proposed that a frailty assessment unit would achieve better outcomes for frail Patients, through providing specialist staffing and a more suitable environment.

A discrete event simulation was initially created to identify what would be the number of beds required for this frail cohort. This model was developed to include all emergency medical admissions (excepting patients on stroke or cardiac pathways)

The model allowed data to be captured relating to what would be the average bed occupancy on both the proposed frailty assessment unit alongside the existing Medical Assessment Units and what percentage of these patients would meet the 4 hour ED Standard.

Findings

An 18 Bedded Frailty Assessment unit (FAU) would have an average bed occupancy of 70% allowing for 99% of patients who would suit admission to FAU to be admitted, with 94% of patients waiting for a bed on FAU waiting less than four hour in ED.

A 16 Bedded FAU would have an average bed occupancy of 77% of the time but the number of patients who would breach the four hour ED standard would increase by 8%

Impact

Proposals have been made relating to reconfiguring wards working with the number of beds suggested in the model as a basis.

Additional pieces of work have been performed to solely model the size of the Medical Assessment unit with.

Discussions for how to improve the bed occupancy whilst maintaining the bed occupancy taken place and have involved allowing expected short stay medical patients to outlie in FAU when occupancy is low.

10-year plan Alignment

SHIFT - Hospital to Community: reducing admissions, shifting frail patients to more appropriate, faster-turnover care setting

H1-1002 — Modelling resources needed to reduce out of area placements in the Mental Health Acute Care pathway

Project ID: H1-1002
Cohort: HSMA 1
Authors: Karl Vile (Devon Partnership NHS Trust)
Organisations: Devon Partnership NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Mental Health, Patient Pathways, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_1/H1_1002_modelling_resources_to_reduce_out_of_area_placements_MH/index.qmd

Abstract

Devon Partnership NHS Trust explored changes to its Urgent Care Pathway to reduce system pressure and avoid out-of-county care. Process mapping and a simulation model identified a need for additional beds. Key findings included high occupancy rates, the necessity to reduce demand and lengths of stay, and the importance of increasing bed numbers to enhance patient care and workforce efficiency.

Project description

The research question for Devon Partnership NHS Trust was ’How would Devon Partnership NHS Trust need to change its Urgent Care Pathway in order to reduce pressure in the system and send no patients out of county for care.’ Process mapping was undertaken to understand the system and 3 years of anonymous patient level information (from October 2013 to September 2016) was collected on the care pathway, from referral to discharge.

A simulation model was developed and a number of scenarios were tested to assess the impact on the system of changes in demand for beds, lengths of stay and delayed discharge, and the number of inpatient beds in the system.

There were a number of key findings, which are described later in this paper, including :

There is a pressure in the mental health urgent care system, equivalent to 47 beds. This pressure is managed by purchasing out of area beds and also by running the urgent care system above desired capacity. This means that wards routinely run at above 100% occupancy ; place of safety and extra care areas are used to hold patients whilst waiting for a bed ; leave beds can be backfilled and crisis house beds are spot purchased in some localities.
Both the out of area spend and the extent to which the existing capacity is running ‘hot’ needs to be addressed to reduce the pressure in the system. This will enhance care for patients and the workforce.
There is no one solution and a reduction in demand, lengths of stay (including delayed discharge) combined with an increase in the number of beds is required.
There are not enough beds in the system.
DPT inpatient wards have relatively high occupancy levels, often above 100%, and low lengths of stay. This provides high throughput per bed compared to alternatives, such as step down or out of area beds. As a consequence closing any DPT beds could result in a higher cost alternative.

10-year plan Alignment

SHIFT - Hospital to Community: keeping patients in local/in-county care rather than displaced out-of-area beds. Keeping patients closer to their support networks during crisis allows patients to recover faster and return to their homes/community-based-care sooner.

H1-1003 — Modelling the impact of various strategies to improve weekend discharge rates

Project ID: H1-1003
Cohort: HSMA 1
Authors: Ryan Hunneman (University Hospitals Plymouth NHS Trust)
Organisations: University Hospitals Plymouth NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Discharge, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_1/H1_1003_modelling_impact_strategies_to_improve_weekend_discharge/index.qmd

Abstract

Ensuring continuous inpatient bed availability is crucial for emergency and non-elective care. Discharge rates vary. Process mapping and Discrete Event Simulation (DES) identified bottlenecks and improved discharge pathways. Changes led to increased weekend discharges and better communication, positively impacting the discharge process, especially for complex patients.

Project description

The problem

Ensuring the continuous flow of inpatient bed availability for emergency and non-elective beds within the Acute Hospital setting is critical in enabling access to specialist and emergency treatment for patients. Variation in discharge rates across the seven-day week impact on the availability of inpatient beds both at the weekend and the start of the working week; typically weekends and Mondays are the worst performing days for discharge – this creates operational issues each week and has a negative impact on access to services and key operational performance standards.

Method

Initial stakeholder mapping and analysis
Creation of project teams for Complex Discharges and Weekend Discharges
Process mapping of multiple discharge pathways for simple and complex discharges
Analysis of critical steps to discharge including key delays and process step timings
Modelling of discharge pathways using Discrete Event Simulation (DES) software

Results

The modelling process of identifying discharge pathways, key staff groups required, delays to process steps and the communication flow through the discharge process both during the week and at the weekend was very beneficial to identifying improvements in the discharge process. Whilst the overall process was very complicated to model using DES, identifying which bottlenecks were evident and visualising the entire process was very valuable.

Impact

The changes made as a result of the modelling process and through the strategy groups formed to implement changes have seen positive results; this has contributed to increased discharges at the weekends and improved communication flows through the discharge process which has had a very positive effect on the whole discharge process especially for complex patients.

10-year plan Alignment

SHIFT - Hospital to Community: faster discharge supports flow out of hospital and back into community settings

H1-1004 — Modelling the impact of regionalising cardiac arrest centres

Project ID: H1-1004
Cohort: HSMA 1
Authors: Jess Lynde (South Western Ambulance Service NHS Trust (SWAST)); Hannah Trebilcock (South Western Ambulance Service NHS Trust (SWAST)); Sarah Black (South Western Ambulance Service NHS Trust (SWAST))
Organisations: South Western Ambulance Service NHS Trust (SWAST)
Techniques: Geographic Modelling
Application areas: Cardiology, Discharge
Public code repository: No
Source file: previous_projects/hsma_1/H1_1004_modelling_impact_regionalisation_cardiac_arrest_centres/index.qmd

Abstract

Survival to hospital discharge after out-of-hospital cardiac arrest (OHCA) is around 7-8%. Analysis of the SWASFT Cardiac Arrest Registry showed survival rate variations due to different hospital services. A regional care system could improve survival rates. A collaborative project between ambulance services and hospitals aims to standardise care, addressing gaps and enhancing outcomes through shared data and best practices.

Project description

Problem

Survival to hospital discharge in patients suffering an out of hospital cardiac arrest (OHCA) has remained nationally at around 7-8% for at least the last 6 years. Anecdotal evidence suggests that patients are not always conveyed to their nearest cardiac arrest centre. The analysis of the SWASFT Cardiac Arrest Registry (2015-2016) identified large variation in survival rates for patients conveyed to different hospitals within the South West, possibly due to variation in the services provided.

Current guidelines suggest implementing a regional system of care for OHCA to improve survival rates and neurological outcomes; nonetheless it remains necessary to understand the impact of such a system on journey times and patient survival.

Method

Literature review of the effect of longer travel times on cardiac arrest survival rates
Identify the current OHCA patients bypassing their nearest cardiac arrest centre
Develop a binomial logistic regression to predict survival rates from patient characteristics, travel time and hospital attended
Model the impact of different regional systems of care using a geographical model in Excel
Report initial findings at a stakeholder meeting

Results

The analysis of the SWASFT Cardiac Arrest Registry confirmed that 13% of OHCA patients were bypassing their nearest cardiac arrest centre. The binomial logistic regression identified that patient characteristics (age, gender, UTSTEIN group) and hospital attended had a significant effect on survival rate. Travel time did not affect survival, although the confidence of this effect is limited to the range of travel times included in the available data (maximum 149 minutes).

The geographical model suggests improvements to survival rate are expected from a regional system of care. This is supported by

the differences in survival rate between hospitals cannot be explained by differences in patients or travel distances alone, and
differences in travel times, within the ranges available in the data, do not appear to affect outcomes of patients.

As with all models there are limitations associated with working assumptions and available data.

Impact

The sharing of patient data in a collaborative project between ambulance service and regional hospitals would be unprecedented and highly beneficial in understanding regional variations, and perhaps working towards standardization of care. Understandably, this project has been met with great interest by stakeholders, and they are driving a collective effort to gather data. In addition, the model has already identified gaps in the current standards of care and highlighted the potential gains that could be made from a standardised best practice across all cardiac arrest centres.

H1-1005 — Modelling the potential impact of having a Clinical Decision Unit

Project ID: H1-1005
Cohort: HSMA 1
Authors: Alaric Moore (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Emergency Departments, Reducing Backlogs, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_1/H1_1005_modelling_potential_impact_of_clinical_division_unit_at_RD&E/index.qmd

Abstract

A Clinical Decision Unit (CDU) at the Royal Devon & Exeter NHS Foundation Trust (RD&E) would manage ED patients needing up to 12 hours of care. Analysis showed that a 24/7 CDU with 4 cubicles could handle 2% of attendances, improving the 4-hour standard by 1.7%.

Project description

A Clinical Decision Unit (CDU) is a designated area for Emergency Department (ED) patients who require testing, treatment, and observational medical management for up to 12hrs. Only patients that meet well-defined criteria where there is an expectation that the patient should be able to safely return home and avoid hospital inpatient admission should be transferred to a CDU.

The ED at the Royal Devon & Exeter NHS Foundation Trust (RD&E) is one of the only departments in England seeing over 100,000 patients a year without a CDU.

This project sought to identify the optimum size and impact of a potential CDU.

The method

To understand the patient journey through the department, a pathway map was first produced with the assistance of a multi-disciplinary ED team. In order to understand trends and variation in patient demand, length of stay and potential CDU appropriateness; detailed analysis of anonymised data recorded on the Trust ED system was undertaken.

The ED clinical and managerial team determined that the criteria for CDU inclusion. A model of the ED pathway was built in Simul8 incorporating the CDU clinical acceptance model and actual data from the whole of 2016.

The results

With a CDU open 24hrs a day 7 days a week, 99.9% of the time 4 cubicles would be adequate.

In a year, just under 2% of attendances would be appropriate to route to the CDU which would have an impact of improving RD&E performance against the 4hr standard by +1.7%.

These results will be integrated into the ED capital programme business case going to the RD&E Board in June 2017.

10-year plan Alignment

SHIFT - Hospital to Community: reducing admissions for patients who can be safely managed in ED and discharged the same day with the right tests and clinical mix available, allowing them to return home sooner.

H2-2001 — Using modelling to understand the delays in mental health services in Devon

Project ID: H2-2001
Cohort: HSMA 2
Authors: Simon Wellesley-Miller (Devon Partnership NHS Trust)
Organisations: Devon Partnership NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Mental Health, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_2/H2_2001_understanding_delays_in_MH_in_Devon/index.qmd

Abstract

This project aimed to support Crisis Cafes, offering an alternative service for those in a mental health crisis. Using modelling and data science, it will look at activity diversion, capacity needs, demand, and resource allocation. Seven models for Barnstaple, Exeter, and Torquay were created, demonstrating system demand and capacity, aiding resource planning and financial decisions.

Project description

The aim of this project was to look at the resourcing to support the establishment of Crisis Cafes - an alternative place for people who are in a mental health crisis to go rather than utilising out of hours, emergency or inpatient services.

Crisis Cafes provide increased choice and additional capacity to meet the needs of those in mental health crisis and provide early intervention which could avoid escalation of mental health needs.

The aim of this project was to try to answer the following questions using modelling and data science approaches : - How much activity can be diverted to the cafés? - What capacity will be needed to achieve a robust service? - What and where is the demand? - How to match demand and capacity - Given the limited resources available how many sites to commission and how to allocate, - How to determine suitable catchment areas?

Seven versions of model were created, for the localities of Barnstaple, Exeter, and Torquay. Separate data and models were built for each locality for weekdays and weekends. Pathways were modelled for each service.

The model helped demonstrate the demand and capacity of the system, allowing for appropriate resources to be planned to support the cafes, and associated financial decisions to be made.

10-year plan Alignment

SHIFT - Hospital to Community: modelling demand and capacity for Crisis Cafes as a community-based alternative to out-of-hours, emergency or inpatient mental health services, diverting activity away from acute settings and into local, accessible provision.

H2-2002 — RCHT Eldercare workforce mapping to identify needs and alternatives to cover the frailty pathway

Project ID: H2-2002
Cohort: HSMA 2
Authors: Claire Westoby (Royal Cornwall Hospitals NHS Trust)
Organisations: Royal Cornwall Hospitals NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Frailty, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_2/H2_2002_RCHT_eldercare_worksforce_mapping_frailty_pathway/index.qmd

Abstract

After hospital admission, 12% of people over 70 experience reduced daily living abilities. Those with deteriorated balance and mobility in the first 48 hours had a 17.1% relative risk of death within 14 days. This project aimed to determine frailty workforce placement to impact more patients, reduce length of stay, and improve care quality. The model simulated patient pathways involving frailty nurses and doctors.

Project description

After a hospital admission, 12% of people over 70 experience a reduction in their ability to undertake daily living activities. Older people who saw a deterioration in their balance and mobility in their first 48 hours of hospital admission had a relative risk of death within fourteen days of 17.1%

The aim of this project was to determine where to base the existing frailty workforce in order to impact as many patients as possible and reduce their length of stay, as well as improving the quality of care and patient experience?

The model simulated the pathway from patient arrival, being seen by frailty nurse, and then either being admitted or discharged. Those that were admitted were then seen by a frailty doctor.

10-year plan Alignment

SHIFT - Hospital to Community: optimising frailty nurse and doctor placement to reduce length of stay and the decline in daily living abilities associated with hospital admission, supporting faster, safer return home for older patients.

H2-2003 — Artificial Ian – can a machine learning tool learn when it is necessary to cancel surgery? And can we use geographic modelling methods to improve outpatient clinic utlilisation in the community?

Project ID: H2-2003
Cohort: HSMA 2
Authors: Simon Philpott (University Hospitals Plymouth NHS Trust)
Organisations: University Hospitals Plymouth NHS Trust
Techniques: Machine Learning
Application areas: NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Hospital to Community, Surgery
Public code repository: No
Source file: previous_projects/hsma_2/H2_2003_ML_outpatient_client_utlilisation/index.qmd

Abstract

This project aimed to use AI approaches to improve surgical cancellation decisions, addressing poor patient experiences. The model struggled due to insufficient data. A second project used geospatial visualisation (QGIS) to assess community outpatient clinic demand, finding increasing waiting lists and poor clinic utilisation, leading to improved patient booking into Peripheral Sites.

Project description

On the day surgical cancellations can offer a poor patient experience - there can be multiple cancellations for some patients, and in early 2018 University Hospitals Plymouth NHS Trust ranked as one of the worst for cancellations. The aim of this project was to explore whether AI approaches could be used to create a model that can inform decision making around cancellations, sooner.

There were three key questions for the Machine learning model to answer: - Should I cancel a patient today? - How many patients should I cancel? - Which patient/s should I cancel?

A model was trained but performance was poor – there was insufficient data to allow the model to be able to replicate the human decision making taking place.

As a second project, the Trust used geospatial visualisation approaches (using QGIS software) to explore demand for community outpatient clinics. The work found that there were increasing waiting lists and poor utilisation of clinics. These results were used to improve their ability to book appropriate patients into Peripheral Sites.

10-year plan Alignment

BIG BET - AI to drive patient power and productivity: using a machine learning model to inform same-day decisions on surgical cancellations, aiming to reduce poor patient experience from late, avoidable cancellations.

SHIFT - Hospital to Community: using geospatial visualisation to improve utilisation of community-based outpatient clinics and Peripheral Sites, shifting appropriate patients away from the main hospital site.

H2-2004 — Reducing the delays to glaucoma treatment at Torbay Hospital

Project ID: H2-2004
Cohort: HSMA 2
Authors: Maria Moloney-Lucey (Devon County Council); Hayley Lewis (NHS South, Central and West CSU)
Organisations: NHS South, Central and West CSU
Techniques: Discrete Event Simulation (DES)
Application areas: NHS 10-year plan shifts: Sickness to Prevention, Opthalmology
Public code repository: No
Source file: previous_projects/hsma_2/H2_2004_reducing_delays_glaucoma_treatment_Torbay/index.qmd

Abstract

This project assessed if the Glaucoma pathway increased treatment delays and blindness risk. The model showed reducing visual field check times had little impact on waiting times, but adding a second Photo & Scan machine nearly eliminated queues for this part of the pathway.

Project description

The aim of this project was to use modelling approaches to assess whether the current configuration of the Glaucoma pathway at Torbay Hospital was increasing delays in treatment, and thus increasing the risk of blindness.

The modelling found that reducing the amount of time visual field checks took (which were perceived to be a bottleneck) had little predicted impact on waiting times. The model also found that adding a second Photo & Scan machine virtually eliminated queues for this part of the pathway.

10-year plan Alignment

SHIFT - Sickness to Prevention: reducing delays in the glaucoma treatment pathway to lower the risk of preventable, irreversible vision loss through earlier and timelier intervention.

H2-2005 — Using modelling to optimise catheter lab efficiency and predict stroke bed demand

Project ID: H2-2005
Cohort: HSMA 2
Authors: Richard Barrett (Royal Cornwall Hospitals NHS Trust)
Organisations: Royal Cornwall Hospitals NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Hospitals, Demand & Capacity
Public code repository: No
Source file: previous_projects/hsma_2/H2_2005_optimise_catheter_lab_efficiency_predict_stoke_bed_demand/index.qmd

Abstract

This project supported a business case for reconfiguring Acute Stroke Unit beds using a Discrete Event Simulation model. The model assessed patient cohorts, pathways, and bed requirements. It also aimed to improve Cardiac Catheter Labs’ efficiency, addressing a 13% capacity loss and reducing patient delays.

Project description

This project was to be used to support a business case being prepared for reconfiguring beds on the Acute Stroke Unit in line with the Peer review recommendations

A Discrete Event Simulation model was developed to represent the entire pathway. The model took into account four different cohorts of patients and based on historic activity data, applied Inter arrival time distributions to each cohort, branched patients proportionally through the various pathways and applied variable lengths of stay dependant on diagnosis and discharge destinations. The model included a range of parameters that could be modified between simulations to assess various scenarios and their impact on bed requirements.

The project also explored using modelling to improve the efficiency of the Cardiac Catheter Labs. The trust had two labs each running two scheduled 4 hour lists per day Mon-Fri, but for numerous reasons the lists were not being utilised efficiently with consequential impacts on both inpatient and elective time from referral to procedure. During 2017 this resulted in an average list duration of 208 minutes representing a 13% loss of capacity. If this lost capacity could be eliminated/reduced this would: • Reduce overall length of stay for emergency patients waiting for lab procedures which in turn will help with overall hospital flow. • Improve patient pathways • Tackle patient delay for elective diagnostics/interventions which is in line with core Trust Objectives.

Meetings with the booking teams enabled the creation of a process flow diagram which highlighted a convoluted process from initial referral to eventual scheduling of procedures.

H2-2006 — Improving ambulance response times for life-threatening emergencies using simulation modelling

Project ID: H2-2006
Cohort: HSMA 2
Authors: Fiona Willmott (South Western Ambulance Service NHS Trust (SWAST))
Organisations: South Western Ambulance Service NHS Trust (SWAST)
Techniques: Discrete Event Simulation (DES)
Application areas: Ambulance
Public code repository: No
Source file: previous_projects/hsma_2/H2_2006_improving_ambulance_response_times/index.qmd

Abstract

Category 1 ambulance calls are life-threatening, making up 8% of calls in the South West in 2018. The project aimed to reduce response times by reserving resources for these calls. A discrete event simulation model was created to determine the optimal resource allocation, balancing the needs of high and low acuity patients.

Project description

Category 1 ambulance calls are the most serious. They are immediately life threatening and, in 2018, approximately 8% of calls in the South West fell into this category. After the triage the call is passed to the dispatchers who co-ordinate the available resources and aim to meet these response times. The response time is the time from the call being made to the arrival of an appropriate ambulance resource on scene. At the beginning of 2018, the ambulance trust was not meeting these times and our number one priority was to reduce the time taken to get to category 1 calls.

The aim of this project was to explore what would happen if dispatchers reserved a portion of resources for the higher acuity calls. Specifically : • What is the optimum level of resources to “hold back” in order to meet the needs of category 1 patients? • What impact would this have on the lower acuity patients? • What is the optimum balance between these two?

A discrete event simulation model was built which could specify a certain number of resources to reserve for category 1 calls. When a call comes in, if it is category 1 call it is allocated an ambulance as soon as one is available. If it is any of the other lower priority categories, an ambulance will only be allocated if this will leave at least the reserved number of ambulances available.

H2-2007 — Can ambulance dispatch codes be used to determine when an ambulance is really needed?

Project ID: H2-2007
Cohort: HSMA 2
Authors: Jessica Lynde (South Western Ambulance Service NHS Trust (SWAST))
Organisations: South Western Ambulance Service NHS Trust (SWAST)
Techniques: Discrete Event Simulation (DES)
Application areas: Ambulance, NHS 10-year big bet: AI to drive productivity
Public code repository: No
Source file: previous_projects/hsma_2/H2_2007_can_ambulance_dispatch_codes_determine_if_really_needed/index.qmd

Abstract

This project aimed to rank dispatch codes by clinical acuity using Machine Learning, incorporating PPI group insights on medical history and co-morbidities. Phase 1 used call-taker info and Logistic Regression to optimise accuracy, leading to phase 2 algorithm generation. The active risk stratification model increased HART team utilisation, leading to a trial of Enhanced Hear and Treat.

Project description

The aim of this project was to rank dispatch codes by typical clinical acuity of the patients in those codes using Machine Learning approaches. The project incorporated the views of a PPI group who helped to highlight the importance of patient medical history and co-morbidities in the model.

Phase 1 of the model used call-taker info (dispatch code and other items) plotted against conveyance. A series of Logistic Regression models was trained to test out optimal sample sizes and model regularisation, to boost accuracy. This led to Algorithm generation for phase 2, to test whether the scoring methodology in the Risk Stratification model could locate codes ‘not requiring an ambulance’.

The risk stratification model is in active use, increasing HART team utilisation, and leading to a trial of Enhanced Hear and Treat. The project allowed for a balanced pragmatic approach by senior decision makers.

10-year plan Alignment

BIG BET - AI to drive patient power and productivity: using machine learning and risk stratification on dispatch codes to identify when an ambulance is not clinically required, supporting smarter resource allocation and Enhanced Hear and Treat as an alternative to conveyance.

H2-2008 — Reducing waiting times for spinal patients using simulation modelling

Project ID: H2-2008
Cohort: HSMA 2
Authors: Claire White (Somerset NHS Foundation Trust); Laura Copp (Somerset NHS Foundation Trust); Ramya Peter (NHS England)
Organisations: NHS England, Somerset NHS Foundation Trust
Techniques: Machine Learning
Application areas: Reducing Backlogs
Public code repository: No
Source file: previous_projects/hsma_2/H2_2008_reducing_waiting_times_spinal_patients/index.qmd

Abstract

Routine reporting revealed increasing wait times for spinal patients, with elective patients waiting too long for treatment decisions due to factors like more appointments, diagnostic tests, and a complex pathway. A discrete event simulation of the spinal pathway looked at referral routes and wait times, showing variation in patient queues per consultant for further investigation.

Project description

Routine reporting exposed a continuing increase in the time spinal patients were waiting, and elective spinal patients were waiting too long for a decision to be made on their treatment plan. There were many contributing factors such as increased appointments, diagnostic tests and a complicated pathway

A discrete event simulation of the spinal pathway was developed, which looked at the referral routes in and wait times between appointments. The model demonstrated some variation across the number of patients in the queues per consultant for further investigation.

H2-2009 — Testing interventions to reduce the Continuing Healthcare Assessment backlog

Project ID: H2-2009
Cohort: HSMA 2
Authors: Ryan Worth (NHS South, Central and West CSU)
Organisations: NHS South, Central and West CSU
Techniques: System Dynamics
Application areas: Reducing Backlogs
Public code repository: No
Source file: previous_projects/hsma_2/H2_2009_testing_intervention_to_reduce_continuing_healthcaare_assesessment_backlog/index.qmd
Links:
- Code: https://insightmaker.com/insight/6eo34fBaxCxf6Grx6KFCZg/CHC-Model-v1-RW

Abstract

A System Dynamics model was developed to show the rated of system flow and interdependencies of components in the system. The model aids Clinical Commissioning Groups planning and decision-making, supporting future demand management.

Project description

In the first quarter of 2017/18 the number of people waiting longer than 28 days for their Continuing Healthcare (CHC) Assessment was 54.3 per 50,000 compared to the national average of 10.2 per 50,000.

A System Dynamics model was built to show the rates of system flow and interdependencies of components in the system.

The model found that : • There were 171 waiting for Decision Support Tool (DST), 51 waiting for Decision and 1009 CHC Eligible. • Numbers waiting for DST were predicted to reach controlled levels during Quarter 1 of 19/20. • A reduction in numbers awaiting Decision was only predicted to occur following a complete reduction in those waiting DST.

The model was made available freely using InsightMaker - CHC Model v1 RW | Insight Maker

Quote from Programme Manager: “The model provides a visual understanding of our waiting list and supports CCG infrastructure and resource planning for future levels of demand. It supports commissioning decision making and helps to focus the team in specific areas within the system.”

H2-2010 — Developing a model to understand delays to discharge

Project ID: H2-2010
Cohort: HSMA 2
Authors: Rohan Kandasamy (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Discharge, Hospitals, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_2/H2_2010_develop_model_understand_delays_to_discharge/index.qmd

Abstract

The hospital faces increasing bed pressures, with over 100 medically fit patients often waiting for discharge due to social care and rehabilitation delays. This project aimed to create a Discrete Event Simulation model to estimate patient length of stay based on demographics, comorbidities, and social care needs, experimenting with scenarios like reducing placement times and accelerating discharge for frail patients.

Project description

The hospital is under ever-increasing bed pressures; at any one time, over 100 medically fit patients can be waiting for discharge. Some of this delay is due to provision of social care, some of this delay is due to ongoing rehabilitation (physio/occupational therapy workup).

The aim of this project was to create a Discrete Event Simulation model that attempted to estimate an individual patient’s length of stay (LOS) based on patient demographics, comorbidities and previous admission history, as well as the patient’s social care need.

The model experimented with various scenarios, including reducing placement times, and accelerated discharge for frail patients.

10-year plan Alignment

SHIFT - Hospital to Community: modelling individual length of stay and social care needs to accelerate discharge for medically fit and frail patients, reducing unnecessary time spent in hospital once treatment is complete.

H3-3000 — What are they saying about us? An AI tool to determine the sentiment of tweets to police forces across the country, and what people are talking about

Project ID: H3-3000
Cohort: HSMA 3
Authors: Mike Hill (Avon & Somerset Constabulary)
Organisations: Avon & Somerset Constabulary
Techniques: Sentiment Analysis, Natural Language Processing (NLP)
Application areas: Police
Public code repository: No
Source file: previous_projects/hsma_3/h3_3000_ai_sentiment_tweet/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=Vj2a-Q_RwKo&list=PLgHO2TgIJXdnAWY7tGGPQLWYAjn0tgG_A&index=6

Abstract

This project used AI-based Natural Language Processing to develop a dashboard predicting tweet sentiment and identifying discussion topics for police forces. It transformed Avon and Somerset Constabulary’s social media can respond to public concerns.

Project description

This project used AI-based Natural Language Processing methods to develop a dashboard that predicts the sentiment (positive or negative) of every tweet to every police force in the country, and automatically identifies the topics that people are talking about positively or negatively. The project has transformed the way in which Avon and Somerset Constabulary’s social media team can respond to concerns identified by members of the public. The HSMA who led this project is now mentoring a project in the fourth round of the programme, to help spread the skills and knowledge he acquired during the programme.

H3-3001 — Generating a richer understanding of relationships in crime data in order to identify opportunities to safeguard individuals and families

Project ID: H3-3001
Cohort: HSMA 3
Authors: Jenna Thomas (Devon & Cornwall Police); Charly Bartlett (Devon & Cornwall Police); Neil Mitchell (Devon & Cornwall Police)
Organisations: Devon & Cornwall Police
Techniques: Network Analysis
Application areas: Police, Crime
Public code repository: No
Source file: previous_projects/hsma_3/h3_3001_crime_data_safeguarding_network/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=oSPxFaeDa5Q&list=PLgHO2TgIJXdnAWY7tGGPQLWYAjn0tgG_A&index=5

Abstract

This project used Network Analysis to identify social relationships and enable proactive police interventions for those at risk of offending due to their links with offenders. A proof of concept showed it could identify “at risk” individuals in hours instead of months. Applied in Devon, it highlighted significant benefits for future policing, including child protection and resource savings.

Project description

This project used Network Analysis methods to identify the social relationships between groups of people who could be causing significant issues within their communities, and enable the police to consider pro-active interventions to target those who may be at risk of future offending because of their social links with offenders. The team developed a proof of concept initially that demonstrated they could use this type of analysis to identify “at risk” individuals in a matter of hours compared to over many months using traditional manual methods. The methods were then applied to identify a highly connected network of youths causing issues in a community within a Devon city.

Fiona Bohan, Performance and Analysis Manager at Devon and Cornwall Police, had this to say about the work :

“The need for Social Network Analysis is clear. Criminal and exploitative networks are a huge and costly issue for police, their partner agencies, and the community. SNA, using minimal resource, can identify children at current and future risk of exploitation, as well as those key players within the network who pose the greatest risk. By targeting and removing these key players, and concentrating limited early intervention resources on protecting those at risk in the future, there are potential significant savings both in terms of child harm and partnership spend. This proof of concept has highlighted the significant benefits of embedding this approach in force and, in my view, will play an important part in policing in the future.”

H3-3002 — Modelling strategies to reduce the elective backlog in hip surgery

Project ID: H3-3002
Cohort: HSMA 3
Authors: Claire Rudler (NHS Devon ICB); Matt Smith (NHS Devon ICB); Imca Hensels-Pelling (Royal Devon University Healthcare NHS Foundation Trust); Zoe Ficken (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust, NHS Devon ICB
Techniques: Discrete Event Simulation (DES)
Application areas: Elective Surgery, Orthopaedics
Public code repository: No
Source file: previous_projects/hsma_3/h3_3002_elective_backlog_hip/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=VLituG9Sbc0&list=PLgHO2TgIJXdnAWY7tGGPQLWYAjn0tgG_A&index=4

Abstract

This project used Discrete Event Simulation to model the hip surgery patient pathway in Exeter, assessing strategies to reduce the surgery backlog. It was a collaborative effort between the CCG and the acute provider, contributing to long-term planning and decision-making.

Project description

This project used Discrete Event Simulation to model the hip surgery patient pathway in Exeter to understand the potential impact of various strategies to reduce the backlog for hip surgery. The work represented a collaboration between the CCG and the acute provider and fed into their long-term planning to support decision-making.

H3-3003 — Developing a generic vaccination service model for the COVID-19 pandemic and beyond

Project ID: H3-3003
Cohort: HSMA 3
Authors: Dr Adam Kwiatkowski (Torridge Primary Care Network)
Organisations: Torridge Primary Care Network
Techniques: Discrete Event Simulation (DES)
Application areas: Vaccination, COVID-19, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_3/h3_3003_generic_vaccination/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=Ve5LuePWrQ4&list=PLgHO2TgIJXdnAWY7tGGPQLWYAjn0tgG_A&index=3
- Code: https://colab.research.google.com/drive/1d7AFLHy6XWvz-_l6MPb2U2bOew6lwk5f?invite=CM3zt9AG
- News Story: https://arc-swp.nihr.ac.uk/news/hsma-covid-vaccination-model/

Abstract

In winter 2020, a mass COVID-19 vaccination program required GP surgeries to deliver vaccines at an unprecedented scale. A Discrete Event Simulation model was the vaccination pathway in North Devon, predicting vaccination rates and identifying potential issues. The model was refined for safe delivery and made available for future use, aiding efficient vaccination efforts globally.

Project description

In winter 2020, vaccinations had recently been approved for use in the UK to protect people against the COVID-19 virus. Consequently, a mass vaccination programme was required to vaccinate most of the population of the country, which led to local GP surgeries having to pivot to deliver vaccinations for their communities at a scale and pace never seen before.

This project was led by a HSMA who had never undertaken any coding work before but using his training from the HSMA programme he rapidly developed a Discrete Event Simulation model of the proposed pathway and resourcing for a vaccination service in North Devon. The model was able to predict not only the rate at which patients could be vaccinated with proposed resourcing, but also the potential risks of social distancing breaches in the waiting room and overflowing in the carpark. The model identified potential issues with the proposed plans, and was used to refine the plans to enable a safe but efficient delivery of the vaccinations in North Devon.

The HSMA also worked with the South West Academic Health Science Network (AHSN) to develop a generic version of the model that could be used for any future vaccination services, and has made the model available Free and Open Source for anyone to use anywhere in the world:

Dr Kwiatkowski said: “I enjoyed designing it. It predicts queue lengths, car park capacity and times for every step of the vaccination process in the clinic to avoid overcrowded waiting rooms. In the winter of 2020/1 the country was back in lock down and there seemed no end in sight regarding COVID. The vaccines seemed like a ray of hope and setting up the clinics was instrumental in turning the tide. Being able to use the knowledge gained from the HSMA course to help design the process was fantastic. The clinic has now delivered over 100,000 vaccines.”

10-year plan Alignment

SHIFT - Sickness to Prevention: modelling and refining a mass vaccination pathway delivered through GP surgeries to enable safe, efficient, large-scale preventative immunisation against COVID-19.

H3-3004 — Exploring the use of Machine Learning and Natural Language Processing to teach a machine to predict whether a patient is likely to be imminently admitted to hospital based on GP data and clues in GP notes

Project ID: H3-3004
Cohort: HSMA 3
Authors: Dr Adam Kwiatkowski (Torridge Primary Care Network); Prof Mona Nasser (Peninsula Dental Social Enterprise); Imca Hensels-Pelling (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Torridge Primary Care Network, Peninsula Dental Social Enterprise, Royal Devon University Healthcare NHS Foundation Trust
Techniques: Natural Language Processing (NLP), Machine Learning
Application areas: Inpatient Admissions, Primary Care (GP), Hospitals, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to Deliver Impact
Public code repository: No
Source file: previous_projects/hsma_3/h3_3004_ml_nlp/index.qmd

Abstract

This project used Natural Language Processing to automate key information extraction from patient notes, aiming to predict imminent hospital admissions. Findings included indicators like increased note frequency, verbosity, and specific words.

Project description

This exploratory project explored the use of Natural Language Processing methods to automate the extraction of key information from free-text patient notes in order to ascertain whether clues in notes could be used to predict an imminent admission to hospital. The team found certain important aspects that seemed to offer indications of an imminent admission, such as increased frequency and verbosity of patient notes, and the use of certain words. The team is now continuing to use these preliminary insights to develop a machine learning algorithm to try to predict an imminent admission for patients in Devon.

10-year plan Alignment

BIG BET - Data to deliver impact: extracting structured signals from unstructured free-text GP notes to build a predictive model, using routine primary care data as the raw material for identifying patients at risk of admission.

SHIFT - Sickness to Prevention: aiming to predict imminent hospital admissions from clues in GP records, enabling pre-emptive intervention in primary care before a patient’s condition deteriorates to the point of needing hospital care.

H3-3005 — Simulation modelling to test proposed models of pediatric critical care

Project ID: H3-3005
Cohort: HSMA 3
Authors: Adam Scull (NHS England)
Organisations: NHS England
Techniques: Discrete Event Simulation (DES), Mapping, Location Optimization
Application areas: Paediatric, Intensive Care Units & Intensive Therapy Units (ICU & ITU), Inpatients
Public code repository: No
Source file: previous_projects/hsma_3/h3_3005_simulation_paediatric_critical_care/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=eOqujR64JkQ&list=PLgHO2TgIJXdnAWY7tGGPQLWYAjn0tgG_A&index=1

Abstract

This project used Discrete Event Simulation and Geographic Modelling to identify optimal locations for Level 2 Paediatric Critical Care Units in South West England, highlighting the need for units outside Bristol. Proposed locations were identified to better serve the population.

Project description

This project used a combination of Discrete Event Simulation and Geographic Modelling approaches to determine where Level 2 Paediatric Critical Care Units should be located in South West England to bring care closer to home. The project demonstrated the clear need for units outside of Bristol to better serve the population in the South West, and was able to identify proposed locations for such sites.

The HSMA who led this project is now mentoring a project in the fourth round of the programme, to help spread the skills and knowledge he acquired during the programme.

H4-4000 — The Effect of Booked Appointments on Waiting Times at Urgent Treatment Centres

Project ID: H4-4000
Cohort: HSMA 4
Authors: Alice Waterhouse (NHS England)
Organisations: NHS England
Techniques: Discrete Event Simulation (DES)
Application areas: Urgent Treatment Centres (UTC), Emergency Departments, ECDS Dataset, Booked Appointments in Walk-in Services, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_4/h4_4000_booked_appointments_utc/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=W1cK5vVukXc&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=6

Abstract

The project aimed to determine if booked appointments reduce waiting times in Urgent Treatment Centres. Using a Discrete Event Simulation model, it showed that higher percentages of booked appointments led to decreased waiting times.

Project description

The aim of the project was to examine in more detail the question of whether booked appointment times might reduce waiting times in Urgent Treatment Centres (UTC). This was done by considering a “generic” UTC – a kind of average over services of this type which were submitting high quality data to the Emergency Care Dataset.

Data on the paths patients took through these services, how frequently patients arrived and the number of staff present were taken from this data and used to build a Discrete Event Simulation model. The modelled average time taken for initial assessment of the patient, treatment, and any investigations or tests was chosen to give the best fit to actual waiting times. The distribution of patient arrival times was then modified to simulate the introduction of additional booked appointments, leading to a more even spread of patient arrivals throughout the day.

The model showed a decrease in overall waiting times when higher percentages of patients booked a timeslot in advance. This will be used to better inform policy on the number of booked appointments needed to make a significant impact to waiting times. Further analysis is expected to shed light on the impact of different ways of prioritising patients according to both clinical need and whether they have booked in advance.

This is a great test case for the use of Discrete Event Simulation to inform policy and strategy at a national level. It demonstrates the impact that such modelling can have and creates opportunity for similar work in the future, ensuring that decision making is supported by high quality analytical insight.

10-year plan Alignment

SHIFT - Hospital to Community: supporting patients to book into the most appropriate urgent care service in advance, evening out arrival patterns and reducing waiting times at Urgent Treatment Centres and improving access to centres closer to the patient.

H4-4001 — South East Regional Covid 19 Vaccination Demand & Capacity Modelling

Project ID: H4-4001
Cohort: HSMA 4
Authors: Charlene Black (NHS South, Central and West CSU); Adonis Sithole (NHS South, Central and West CSU); Edward Chick (NHS South, Central and West CSU); Mayoor Dhokia (Surrey and Borders Partnership NHS Foundation Trust)
Organisations: NHS South, Central and West CSU, Surrey and Borders Partnership NHS Foundation Trust
Techniques: Forecasting, Prophet, Plotly Dash
Application areas: Vaccination, COVID-19, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_4/h4_4001_covid_19_vaccination_demand_capacity/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=_AWTVeNqOwM&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=3

Abstract

The project aimed to create an easy-to-use model for predicting Covid vaccination demand and capacity. Despite data access barriers, the team made progress by using past vaccination data. The tool will help track vaccinations, identify problem areas, and save time and money.

Project description

The aim of the project was to create an easy to use model for predicting demand and capacity for Covid vaccinations.

During the project the team came up against many barriers, the biggest of which was getting access to the data. The other issue they had with getting the data was the limits imposed on the amount of data you could download in one go, meaning that it was a lengthy process.

They modelled potential update scenarios and have started to work on this and have making great progress. They will use the previous vaccination data, along with the Winter 2022 vaccination data to strengthen the tool, so that following vaccination seasons it can be used more widely. They will look at other vaccinations it could be applied to, such as flu.

This will give users an easy to use tool to keep track of vaccinations and pick up problem areas quickly. It will save time and money on current methods and enable more people to use and understand the vaccination position

10-year plan Alignment

SHIFT - Sickness to Prevention: forecasting demand and capacity for COVID-19 (and potentially flu) vaccination programmes to support efficient, preventative immunisation delivery across the region.

H4-4002 — Use of Discrete Event Simulation to Tackle Long Waits and a Growing Backlog for Children Requiring Neuro Development Assessment (Autism and ADHD)

Project ID: H4-4002
Cohort: HSMA 4
Authors: Irma Tanovic (Oxford Health NHS Foundation Trust)
Organisations: Oxford Health NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Neurodiversity, Paediatric, Mental Health, Demand & Capacity, Reducing Backlogs, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_4/h4_4002_des_adhd_autism_pathway/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=IuD0tmKtjV4&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=13

Abstract

The project aimed to identify bottlenecks in the Autism and ADHD assessment pathway for children and determine changes needed to manage demand and avoid backlogs. The main issue was clinical staffing capacity. Could adding one clinician could maintain a steady state and prevent backlog growth?

Project description

The aim of the project was to understand the current pathway to assessment of Autism and ADHD in children, identifying any bottlenecks. We wanted to know what change needed to take place in order to keep on top of ongoing demand levels and see patients within an acceptable time - what do we need to do to maintain a steady state and avoid a backlog from developing?

In addition to maintaining a steady state, we wanted to understand how to eliminate the backlog that has developed as a result of the ongoing demand not being met.

The main bottlenecks were identified and the issue is not at the beginning of the pathway (getting the completed questionnaires back from parents/schools) like it has been suggested at the start of the project, but further down the pathway due to clinical staffing capacity. Furthermore, the developed model shows that recruiting just one extra clinician would result in having a steady state with the backlog not growing. The model also shows possible staffing requirements to remove existing backlog in 1-5 years in addition to keeping a steady state.

The findings and the model demo have been presented to the stakeholders including service managers, clinical and operational leads, service heads and the Trust board, and discussions on plans for how the model is used to support any improvement work in the service is already taking place.

10-year plan Alignment

SHIFT - Hospital to Community: identifying staffing needed to clear the backlog and maintain a steady state for children’s autism and ADHD assessments, improving timely access to community and outpatient-based neurodevelopmental services. By providing timely access to diagnosis and resources, we prevent patients escalating to needing more serious, potentially hospital-based, services.

SHIFT - Sickness to Prevention: By providing timely diagnosis, patients with ADHD and autism can be supported to live well, with interventions being provided at an earlier stage and minimising the risk of developing comorbid conditions secondary to the primary diagnosis, such as depression and anxiety.

H4-4003 — Using DES to Improve Flow through an Acute Medicine Assessment Pathway

Project ID: H4-4003
Cohort: HSMA 4
Authors: Helen Young (Nottingham University Hospitals NHS Trust); Thomas Knight (Sandwell and West Birmingham NHS Trust)
Organisations: Nottingham University Hospitals NHS Trust, Sandwell and West Birmingham NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Emergency Departments, Patient Flow, Inpatients
Public code repository: No
Source file: previous_projects/hsma_4/h4_4003_des_amu_flow/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=TDTyBcVZDUU&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=9

Abstract

Nottingham University Hospitals’ Acute Medicine Assessment areas face high occupancy, causing delays in the Emergency Department. A Discrete Event Simulation showed that if patients left within 16 hours, ED wait times over 6 hours would reduce by 90%. This insight will help quantify additional ward beds needed to improve patient flow and reduce harm.

Project description

Nottingham University Hosptials has two major Acute Medicine Assessment areas (WB3 and AMRA), where patients can be reviewed before being sent to the most appropriate specialty base ward bed for their needs.

WB3 and AMRA run at very high occupancy levels, and patients frequently wait in the Emergency Department (ED) for a WB3/AMRA bed for more than 6 hours – this is a known cause of delay-related harm. Stakeholders feel that a major constraint is patients waiting in WB3/AMRA who are ready to be moved to a ward bed.

The team wanted to investigate what the impact would be if patients were able to leave WB3/AMRA more quickly. What would this do to the backlog of queues in ED, and how would it affect waiting times there?

A Discrete Event Simulation (DES) was made of the entire Acute Medicine assessment pathway. They then simulated several alternative scenarios in which patients leave WB3/AMRA more quickly after their period of assessment is completed – and were able to see the impact of this in the key metrics.

If all patients left WB3/AMRA within four hours of their request for a ward bed, patients would never have to wait in ED for a WB3/AMRA bed for more than 6 hours. Even if all patients left WB3/AMRA within 16 hours of their request, this would still reduce the number of patients waiting more than 6 hours by 90% - substantially reducing the risk of delay-related harm

We can now start to quantify the number of additional ward beds that would need to be freed up to support this improved flow, and bring these findings to discussions on how/where additional bed days could be released.

H4-4004 — Discrete Event Simulation of Cognitive Behavioural Therapy Pathway in an IAPT Service

Project ID: H4-4004
Cohort: HSMA 4
Authors: Katie Brown (Dorset HealthCare University NHS Foundation Trust); Hannah Carroll (Dorset HealthCare University NHS Foundation Trust); Andrew Poole (NHS Dorset ICB); Matthew Chapman (Dorset HealthCare University NHS Foundation Trust)
Organisations: Dorset HealthCare University NHS Foundation Trust, NHS Dorset ICB
Techniques: Discrete Event Simulation (DES)
Application areas: Mental Health, NHS Talking Therapies (Formerly IAPT), Community Mental Health
Public code repository: No
Source file: previous_projects/hsma_4/h4_4004_des_cbt_iapt/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=KPmxGzSuN0s&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=4

Abstract

The project used Discrete Event Simulation to model the wait list for High Intensity Cognitive Behavioural Therapy. It found that changing an evening group to a face-to-face clinic could reduce wait times, especially for face-to-face or evening 1-1 appointments.

Project description

The aim of the project was to use Discrete Event Simulation (DES) to model the current wait list for High Intensity Cognitive Behavioural Therapy (HI-CBT). The team wanted to see if changing the allocation of therapist resources by running an evening face-to-face clinic, instead of an evening group, would improve overall wait times for therapy.

Patients were created as entities within the model, and average % preferences for IESO (online therapy), group therapy, virtual or face-to-face 1-1 appointments were assigned. Along with whether the patient was a priority, preferred evening or face-to-face appointments, and average number of appointments offered. They used current and historic data of people waiting or seen for therapy, attendance rates, and meeting with the data lead and service managers within Steps 2 Wellbeing, in order to gain an accurate picture of the waiting and treatment pathway for patients.

The model does confirm that the longest wait times appear to be for people waiting for face-to-face or evening 1-1 appointments. The model shows improvements in waiting times when an evening group is changed to a 1-1 face-to-face evening clinic.

The team is making changes to the model based on stakeholder feedback and will present to senior management within the service.

H4-4005 — Use Of Discrete Event Simulation (DES) to reduce delays in Cancer Diagnosis & Treatment

Project ID: H4-4005
Cohort: HSMA 4
Authors: Angel Masih (Gloucestershire Hospitals NHS Foundation Trust); James Page (University Hospitals of Morecambe Bay NHS Foundation Trust); Mahya Kaveh (Dartford and Gravesham NHS Trust)
Organisations: Gloucestershire Hospitals NHS Foundation Trust, University Hospitals of Morecambe Bay NHS Foundation Trust, Dartford and Gravesham NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Cancer, Colorectal, Identifying Bottlenecks in Pathways
Public code repository: No
Source file: previous_projects/hsma_4/h4_4005_des_delays_cancer_diagnosis_treatment/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=Wmpu4fpu0J4&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=2

Abstract

The project developed a Discrete Event Simulation model for the Colorectal Cancer pathway identify key delays and solutions. Using three years of anonymised patient records, the model highlighted diagnostic delays during investigative stages.

Project description

The project aim was to develop a Discrete Event Simulation (DES) model based on Colorectal Cancer pathway at Gloucester Royal Hospital to identify the key delays within the pathway and identifying feasible solutions.

The team met with cancer service leads to discuss the scope of the project and discussed the patient pathway in detail. Three years of anonymised patient records were used to create the model. Delays for the pathway were split into patient & provider delays to focus only on provider delays.

They used SimPy to create a DES model using the current colorectal patient pathway. They ran the simulation for 100 runs to get the average waiting time for each stage of the pathway to identify the stages that took the longest amount of time. The DES model showed us the key delay, which is diagnostic delays caused during investigative stages.

The findings from the analysis will be presented to cancer leads to support reducing the waiting times for colorectal patients. The DES model helped to evidence key bottlenecks of the patient pathway and will help to discuss the prospect of how increased resource can help improve patient outcomes.

H4-4006 — Discrete Event Simulation to Improve Flow and Performance in the Urgent Treatment Centre

Project ID: H4-4006
Cohort: HSMA 4
Authors: Shilpa Patel (University College London Hospitals NHS Foundation Trust)
Organisations: University College London Hospitals NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Urgent Treatment Centres (UTC), Emergency Departments, Identifying Bottlenecks in Pathways
Public code repository: No
Source file: previous_projects/hsma_4/h4_4006_des_flow_performance_utc/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=3JO8_70hHdA&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=11
- News Story: https://arc-swp.nihr.ac.uk/news/data-improves-urgent-care/

Abstract

The project aimed to improve Urgent Treatment Care by developing a model to help the Emergency Department allocate staff and rooms efficiently. The model tested alternative pathways and identified the need for additional rooms and better-aligned staffing rotas. These changes aimed to reduce bottlenecks and meet the target for patient flow.

Project description

The primary aim of the project was to improve Urgent Treatment Care (UTC) performance by developing a model which would help the Emergency Department (ED) team with the allocation of staff and rooms to match patient flow. The team wanted to understand what further staff and resources would be required for any variations in patient attendance. The ED team wanted to understand how changes in the patient pathway would reduce bottlenecks, for example, what would be the impact of front-loading diagnostics.

Initially, a discrete event simulation model was developed to show how patients currently flow through UTC. After an initial model had been developed, it was used to test the proposed alternative pathways for diagnostics. It was also used to model different numbers of staff and rooms. The model made it easy to understand the impact of changes to the existing provision and what would be most cost-effective.

The ED department was redesigned based on the model findings, which identified a need for additional rooms leading to reconfiguration of the space. It started a review of staffing rotas to see how they could be better aligned to findings from the analysis and future staffing rotas are designed to align with what the model identified was needed to meet the 95% target.

H4-4007 — Forecasting Demand and Length of Stay in the Emergency Department

Project ID: H4-4007
Cohort: HSMA 4
Authors: Lisa Sabir (Sheffield Children’s NHS Foundation Trust)
Organisations: Sheffield Children’s NHS Foundation Trust
Techniques: Machine Learning, Forecasting, Streamlit
Application areas: Paediatric, Length of Stay, Emergency Departments, Inpatients, Staffing Level Optimisation
Public code repository: No
Source file: previous_projects/hsma_4/h4_4007_forecasting_los_ed/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=a7MtCSbTg0I&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=7

Abstract

The project aimed to use forecasting and machine learning to predict Emergency Department arrivals and length of stay. A web-based app was designed to explore trends and suggest improvements. The app forecasts attendances and stay durations, helping plan workforce needs.

Project description

The project aimed to use forecasting and machine learning methods to predict the length of stay and arrivals to the Emergency Department (ED) at different times of day, days of the week and months of the year. Understanding this would allow the team to examine trends and predict future demand so they better plan workforce.

They have designed a web-based app where they can input the Sheffield Children’s hospital data and use it to explore trends in attendances and length of stay. This allows them to see patterns and suggest improvements, such as when it would be better for extra staffing. The web-page produces a “Forecast” demonstrating the attendances by hour/day/month/year as well as length of stay via interactive buttons. It is then able to give an estimate for a selected time to predict future demand.

They have completed the forecasting methods for this project and are now developing the machine learning model. The team hope to gain permission for this to be used on a wider scale so individual trusts can input their own data.

H4-4008 — Meeting the demand of 111 for primary care services

Project ID: H4-4008
Cohort: HSMA 4
Authors: Richard Pilbery (Yorkshire Ambulance Service NHS Trust); Maddie Smith (NHS Devon ICB); Jon Green (University of Plymouth)
Organisations: Yorkshire Ambulance Service NHS Trust, NHS Devon ICB, University of Plymouth
Techniques: Discrete Event Simulation (DES), Plotly Dash, Network Analysis, Telephone-based Services
Application areas: 111 Service, Primary Care (GP), Emergency Departments, NHS 10-year plan shifts: Hospital to Community
Public code repository: Yes
Source file: previous_projects/hsma_4/h4_4008_meeting_demand_111/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=cg5Bp3O02io&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=1
- Article: https://arc-swp.nihr.ac.uk/publications/modelling-nhs-111-demand-for-primary-care-services-a-discrete-event-simulation/
- Paper: https://bmjopen.bmj.com/content/13/9/e076203
- Code: https://github.com/RichardPilbery/MOOOD-study

Abstract

The project modelled NHS 111 calls triaged to primary care. Simulations showed timely primary care contact could reduce 999 calls and ED attendances but would require nearly doubling primary care services.

Project description

This was an attempt to model the impact of enforcing timely primary care interventions when 111 calls are undertaken.

More information: https://arc-swp.nihr.ac.uk/publications/modelling-nhs-111-demand-for-primary-care-services-a-discrete-event-simulation/

10-year plan Alignment

SHIFT - Hospital to Community: modelling how timely primary care contact following 111 calls could reduce 999 calls and ED attendances, diverting patients toward community-based primary care services.

H4-4009 — Using Machine Learning to Predict Hospital Admissions and Length of Stay for Respiratory Conditions

Project ID: H4-4009
Cohort: HSMA 4
Authors: Andy McCann (NHS Midlands and Lancashire CSU)
Organisations: NHS Midlands and Lancashire CSU
Techniques: Machine Learning
Application areas: Respiratory, Inpatient Admissions, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to deliver impact
Public code repository: No
Source file: previous_projects/hsma_4/h4_4009_ml_admissions_los_respiratory/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=-0AGbib9Ig0&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=14

Abstract

The project aimed to predict respiratory condition admissions and length of stay using patient history and demographics. Early-stage modelling showed promising preliminary findings. Further development will include more patient history to improve predictions and extend the model’s capabilities.

Project description

The aim of the project was to use available Primary and Secondary Care patient history and demographic information to better predict the chance of admission for respiratory conditions and subsequent length of stay, in order to better target interventions.

Due to time and data constraints, the modelling is currently at an early stage, but has demonstrated some interesting preliminary findings and put the structure in place for further development.

A Logistic Regression confirmed some expected factors but also revealed some surprises. An initial neural network with very little optimisation matches the Logistic Regression performance, with the potential to beat it when other features are included.

The model needs to take other features, particularly more patient history, into account to improve performance and be extended to predict length of stay as well as admission.

10-year plan Alignment

BIG BET - Data to deliver impact: combining primary and secondary care patient history and demographic data to build predictive models for respiratory admissions and length of stay.

SHIFT - Sickness to Prevention: aiming to better target preventative interventions for patients at risk of admission for respiratory conditions, informed by predicted likelihood of admission.

H4-4010 — The role of Patient Initiated Follow-up (PIFU) and ‘Digital Outpatients’ in Supporting the Elective Recovery - Can We Better Size Potential for Clearing the Backlog?

Project ID: H4-4010
Cohort: HSMA 4
Authors: Martina Fonseca (NHS England); Xiaochen Ge (NHS England)
Organisations: NHS England
Techniques: Discrete Event Simulation (DES)
Application areas: Patient Initiated Follow-Up (PIFU), Digital Outpatients, Rheumatology, Reducing Backlogs
Public code repository: Yes
Source file: previous_projects/hsma_4/h4_4010_pifu_digital_outpatients_elective_recovery/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=TO5sKaW6BGk&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=12
- Code: https://github.com/nhsx/HSMA4-12-DES-rheum

Abstract

The project aimed to explore the role of Patient Initiated Follow-up (PIFU) in addressing backlogs by mapping rheumatology outpatient pathways using discrete event simulation. The team modelled resource use and redeployment, focusing on PIFU’s impact on referral-to-treatment waiting lists.

Project description

The aim of the project is to find out what role Patient Initiated Follow-up (PIFU) can play in the redeployment of capacity to address the backlog. This will be done by mapping outpatient pathways for rheumatology sub-pathway using discrete event simulation (DES).

The team did what-if modelling of resource use based on the proportion of patients on a PIFU pathway, the rate of PIFU patient-initiated requests and the use of advice & guidance. They used this to understand how released resource is redeployed, and how the upstream referral-to-treatment (RTT) waiting list behaves (size and waiting time to first outpatient).

The focus was on rheumatology as a case study since it had well documented pathways and good clinical evidence base on PIFU, with PIFU actively endorsed by NHSE and GIRFT. It is mainly an outpatient specialty with many chronic patients on long-term follow-up, meaning that the effect of PIFU is in theory amplified.

Other opportunities include discussing a spin-off model that includes effect of other digital musculoskeletal peri-treatment interventions or creating a proof-of-concept generalised DES that can be used for other PIFU specialties (PIFU is being advocated across most elective specialties).

On the PIFU rheumatology model itself, the team aim to continue refining the model based on internal stakeholder feedback. A secondary aim is to create an end-user toy tool that could help demonstrate some of the what-if scenarios to operational managers, in collaboration with NHSE Digital Care Model colleagues.

10-year plan Alignment

SHIFT - Analogue to Digital: using Patient Initiated Follow-up and digital outpatient pathways to redeploy clinical capacity, reducing the need for routine face-to-face hospital appointments for chronic rheumatology patients.

SHIFT - Hospital to Community: helping clear the elective outpatient backlog and supporting hospitals as we know it to end, by shifting follow-up care to patient-initiated, lower-intensity contact.

H4-4011 — Predicting Non-Elective Admissions

Project ID: H4-4011
Cohort: HSMA 4
Authors: Stephen Ashmead (Royal Devon University Healthcare NHS Foundation Trust); Karim Kamara (Royal Devon University Healthcare NHS Foundation Trust); Jahangir Alam (NHS North East London CSU)
Organisations: Royal Devon University Healthcare NHS Foundation Trust, NHS North East London CSU
Techniques: Machine Learning, Natural Language Processing (NLP)
Application areas: Non-elective Admissions, Inpatient Admissions, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: Data to Deliver Impact
Public code repository: No
Source file: previous_projects/hsma_4/h4_4011_predicting_non_elective_admissions/index.qmd

Abstract

The project aimed to develop a predictive model to identify patients at high risk of admission and provide explanatory feedback. It combined structured and unstructured data models. The model helps predict non-elective admissions, enabling preventative care and better health outcomes, despite data limitations.

Project description

The project aim was to develop a predictive model to identify patients at high risk of admission and to provide explanatory feedback as to why the patients were at risk. There were three elements of the project:

Using structured data to generate a predictive model
Use unstructured data from patient notes to generate a predictive model
To combine the two models to create an enhanced model

Three models were developed using structured data – logistic regression, random forest and neural network. Logistic regression performed best.

For the unstructured data, attempts to use more advanced Natural Language Processing techniques (such as neural networks and transformers) were unsuccessful due to limited computing power. However, a more basic model using a Term Frequency-Inverse Document Frequency matrix with a random forest showed some improvement on accuracy.

Electronic health record data can accurately be used to predict whether a patient is at risk of non-elective admission. Administrative events are often better indicators than clinical measures, however EHR data is prone to several limitations and biases which can lead to counter-intuitive correlations. For analysis of unstructured data, greater computing power than I have access to is required to analyse the large quantities of patient notes, even when focusing on relatively short time frames.

It will make a difference to patient care by allowing patients at risk of non-elective admission to receive preventative care, leading to better health outcomes for the individual patients and freeing up inpatient resources for other patients.

10-year plan Alignment

BIG BET - Data to deliver impact: combining structured electronic health record data with unstructured patient notes to build a predictive model of non-elective admission risk.

SHIFT - Sickness to Prevention: aiming to identify patients at high risk of admission so that preventative care can be delivered before deterioration leads to hospital admission.

H4-4012 — Predicting Violent Incidents on Mental Health Inpatient Units

Project ID: H4-4012
Cohort: HSMA 4
Authors: Iain Waring (Birmingham & Solihull Mental Health NHS Foundation Trust)
Organisations: Birmingham & Solihull Mental Health NHS Foundation Trust
Techniques: Machine Learning
Application areas: Mental Health, Mental Health Inpatients
Public code repository: No
Source file: previous_projects/hsma_4/h4_4012_predicting_violent_incidents_mh_inpatients/index.qmd

Abstract

This project aimed to see if violent incidents could be reduced on hospital wards by predicting and preventing them, using machine learning. A dashboard would highlight high-risk wards to senior staff for intervention.

Project description

The aim of the project was to see if the occurrence of violent incidents on hospital wards could be reduced, if incidents could be pre-empted and interventions could be taken to try to prevent them from happening.

The trust has a wealth of electronic information about service users, staff, incidents, and what happens on inpatient units. The plan was to combine this information into a dataset and use machine learning techniques to predict which of our wards were most likely to suffer from an incident the following day. A dashboard would highlight these wards to senior nursing staff, who could experiment with actions to reduce incidents.

Discussions with key staff helped to identify which features from our systems would be the most helpful in predicting incidents. A large dataset was constructed, and machine learning methods were used in an attempt to predict where incidents might happen. Unfortunately, the dataset did not prove to be informative enough to generate a working model – the predictions made were inaccurate and there was no way to apply them to the work environment.

H4-4013 — Reducing Travel Times to Treatment for Cardiac Patients in the South East of England

Project ID: H4-4013
Cohort: HSMA 4
Authors: Glenn Ubly (NHS England Specialised Commissioning – South East Region); Atonia Drummond (NHS England); Janine James (NHS England); Victor Yu (The Strategy Unit)
Organisations: NHS England Specialised Commissioning – South East Region, NHS England, The Strategy Unit
Techniques: Mapping, Travel Times, Streamlit, Location Optimisation
Application areas: Cardiology
Public code repository: Yes
Source file: previous_projects/hsma_4/h4_4013_reducing_travel_times_cardiac_patients/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=28yXjieiMEM&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=6
- Code: https://github.com/GlennUbly/HSMA4_cardiac
- Website: https://glennubly-hsma4-cardiac-introduction-f1fq51.streamlitapp.com/

Abstract

The project analysed travel times to understand the impact of flow of activity into London on patient travel. It identified a gap in cardiac surgery access for Kent & Medway patients, suggesting new sites could help. A Streamlit app visualised these impacts.

Project description

The aim of the project was to employ geographical and statistical analysis of past and present travel times to understand the impact the flow of activity into London has on travel times for patients, whether referral pathways could be changed to minimise patient travel, and the extent to which additional sites would have a beneficial impact.

Initial analysis was undertaken on a sample of cardiac procedures within the hospital spell level data. The analysis identified a particular gap in the accessibility of cardiac surgery for patients in the Kent & Medway area, where there would be a significant benefit for the local population in the provision of a new cardiac surgery site or sites. The methods used gave a quantified and visual view of the impact of the options, and indicated the optimal configurations with 1 or 2 additional sites.

The team created a Streamlit application which allows a user to select from a list of possible new sites, and see the impact on travel times for Kent & Medway patients using maps, charts and a number of key travel time metrics.

The South East Cardiac Network have reviewed the findings, and this work will form part of the evidence base for service planning in the region. More generally, there has been interest in applying the methods and tools to other services.

H4-4014 — Developing a Service Planning Decision Support Tool to Tackle Inequalities and Minimise Carbon Output

Project ID: H4-4014
Cohort: HSMA 4
Authors: Matt Eves (Derbyshire Community Health Services NHS Foundation Trust); Anya Gopfert (Torbay Council); Sally Brown (West Sussex County Council)
Organisations: Derbyshire Community Health Services NHS Foundation Trust, Torbay Council, West Sussex County Council
Techniques: Machine Learning, Automation, Reproducible Analytical Pipelines (RAP)
Application areas: Carbon Emissions, Health Equity Audits, Non-attendance Prediction, NHS 10-year plan shifts: Analogue to Digital, NHS 10-year big bet: Data to Deliver Impact
Public code repository: Yes
Source file: previous_projects/hsma_4/h4_4014_service_planning_inequalities_carbon_output/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=zYWte2obxDs&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=8
- Code: https://github.com/hsma4-student/shine

Abstract

The project explored the feasibility of considering inequalities and carbon emissions in new clinic locations. Achievements include automating a Health Equity Assessment, running logistic regression models to predict appointment no-shows, and estimating patient travel carbon emissions.

Project description

There are three current significant health agendas to which this project broadly relates.

These are

The inequalities agenda
The digital and transformation agenda, and
The Net Zero agenda.

The aim of this project was to explore feasibility of considering the impact of a new clinic location on inequalities and carbon emissions, and any co-benefits or trade-offs between the two. Ultimately, the aim was to enable decision making to incorporate these two significant areas.

The team have achieved:

The automation of a Health Equity Assessment to compare the profile of patients in a service to the population at large
The running of a number of variations to a logistic regression model to identify features predictive of someone not attending their appointment
Quantifying an estimate for the patient travel carbon emissions associated with the current service design.

They are intending to continue working together as a team to finish the project’s initial scope (e.g. inc. possible new clinic sites and model impact on unmet need / emissions) – potentially supported by code from another project, emphasising the benefit of open source.

They have been successful in securing a funding from the Greener NHS Healthier Futures Action Fund joint bid.

10-year plan Alignment

SHIFT - Analogue to Digital: automating a Health Equity Assessment and building a reproducible analytical pipeline to support data-driven, technology-enabled service planning decisions.

BIG BET - Data to deliver impact: combining patient, location and emissions data into a decision support tool to inform new clinic site planning, integrating equity and carbon considerations into service design.

H4-4015 — Spatial Modelling of Violent Crime to Support Strategic Analysis

Project ID: H4-4015
Cohort: HSMA 4
Authors: Anupma Wadhera (TBC); Linda Wystemp (TBC); Andrea Casajuana Massanet (Counter Terrorism Policing); Helen Browne (Devon & Cornwall Police); Alessia Rose (Devon County Council)
Organisations: Counter Terrorism Policing, Devon & Cornwall Police, Devon County Council
Techniques: Mapping, Geostatistics, QGIS
Application areas: Police
Public code repository: No
Source file: previous_projects/hsma_4/h4_4015_spatial_modelling_crime/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=K0g2Iaw3blY&list=PLgHO2TgIJXdn0k4y56xtM-VpcHU_5RJUN&index=10

Abstract

The project aimed to use crime data and spatial analysis for intelligence reporting. Focusing on violent crime, the team used QGIS and Python libraries to analyse geographic offence data. The model identified hotspots, coldspots, and outliers, providing statistical confidence for intelligence reports.

Project description

The aim of this project was to use various crime data variables and spatial analysis to turn them into useful insights for inclusion in assessed intelligence reporting.

The team chose to focus on violent crime offence data as they wanted to create our model using free and open source software.

They used QGIS and various Python libraries to explore the data. For the purpose of their model they obtained detailed geographic offence data uploaded by police forces in England and Wales.

The model was able to take a crime indicator (e.g. violent and sexual offences) for a region in the UK, prepare the data, apply and compare map classification schemes and run various analysis techniques to identify spatial autocorrelation, LISA , identify hotspots, coldspots and outliers. The statistical output provided a level of confidence in the findings that they can incorporate into intelligence confidence reporting levels.

The model will be shared with Commodity Threat Leads with the intention of using it internally to identify spatial trends in datasets that are restricted. Applying spatial analysis to these datasets will help identify regional variation and help quantify visual observations that will provide statistical certainty to any observed visual findings. The model will be proposed and applied to more restricted crime data in order to be used to draw out useful visualisations and insights to feed into intelligence reporting.

H5-5002 — Using Natural Language Processing to detect drug related content within free text

Project ID: H5-5002
Cohort: HSMA 5
Authors: Lauren Szarvas; Holly Dale; Tom Haddock
Organisations: Not recorded
Techniques: Natural Language Processing (NLP), Named Entity Recognition
Application areas: Police
Public code repository: No
Source file: previous_projects/hsma_5/h5_5002_nlp_drug_content/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=7j8_DlMoaSc&t=22s

Abstract

The project aimed to detect drug-related content in datasets using Natural Language Processing to automate text classification. Initial findings were promising, though tested only on dummy data. If successful, the model could be adapted for other crime types.

Project description

Since the amount of data being collected and stored has increased significantly, datasets are often reviewed manually, particularly when free text fields are present. This project focussed specifically on the detection of drug-related content, following the announcement of the Governments 10-year drugs plan in December 2021. If successful, this could then be adapted and applied to further crime types at a later date.

The aim was to create a model that would be able to make predictions on future datasets by classifying each piece of text into drug/non-drug related content, to prevent manual coding of the data. To achieve this, the team used a machine learning technique known as Natural Language Processing, to train various models on the structure of text and any patterns in language.

The team created a model pipeline to automate this process end-to-end. Initial findings suggesting these techniques could be rather successful, however, it has only been possible to test this on dummy data at present. If this were to be implemented as a permanent solution, this could potentially be made available to the wider policing community through the development of a Streamlit app, which could greatly assist in reducing the amount of data and time taken to review each dataset manually.

H5-5003 — Using Discrete Event Simulation to model the bottlenecks in the Acute Medical Unit pathway

Project ID: H5-5003
Cohort: HSMA 5
Authors: Becky Crofts (Royal Devon University Healthcare NHS Foundation Trust); Kayleigh Haydock (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Acute Medical Unit (AMU), Emergency Departments, Acute Care, Hospitals, NHS 10-year plan shifts: Hospital to Community
Public code repository: Yes
Source file: previous_projects/hsma_5/h5_5003_des_bottlenecks_amu/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=_3j0xC_yCKQ
- Code: https://github.com/BeckyCrofts/amu_modelling

Abstract

In this project, a computer simulation of the acute medical pathway in a Devon trust was created, along with an interactive tool allowing parameters such as the staffing levels to be changed. This allowed staff to explore the optimum levels of resourcing, enabling risk-free testing of staffing and resource changes before committing to these changes in the real-world.

Project description

The team’s project was to create a computer model of part of the acute medical pathway in their hospital trust. As part of this process patients come in with non-surgical emergencies, like an abnormal heart beat, and they’re triaged to decide the best route of care for them. In some cases that will be an admission to hospital, but increasingly this may involve alternatives that combine just a few hours in hospital with close follow up and perhaps even forms of remote monitoring, like monitoring their heart through a mobile app.

The aim of this project was to model part of the acute medical pathway in this hospital trust, simulate how the system currently works, and investigate whether there is optimal allocation of staff and resources. The benefit of this model is it allows for changing parameters such as adding extra staffing to see what impact that has on the pathway. Testing these changes in the real-world costs time and money, while computer simulation allows you to trial these changes with minimal cost and no risk.

The model isn’t validated yet, but when it is it’ll be handed over to users (e.g. clinicians and managers) to support them in making decisions about staffing or resourcing. They’ll be able to test those changes mentioned above without risk before proceeding to a real-world test of change.

Access the Github Repository

10-year plan Alignment

SHIFT - Hospital to Community: exploring appropriate resourcing for the AMU to support access to alternatives to admission like remote monitoring, such as monitoring a patient’s heart through a mobile app rather than a longer inpatient stay.

H5-5004 — Modelling the effect of complex discharge delays on acute performance

Project ID: H5-5004
Cohort: HSMA 5
Authors: Hannah Perkins (University Hospitals Plymouth NHS Trust)
Organisations: University Hospitals Plymouth NHS Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Discharge, Hospitals, Patient Flow, Acute Care, Inter-service Interactions, NHS 10-year plan shifts: Hospital to Community
Public code repository: No
Source file: previous_projects/hsma_5/h5_5004_effect_complex_discharge_delays_acute_perform/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=nrUkRha_5F0&t=1s

Abstract

Hospitals face increasing A&E wait times, ambulance delays, and growing waiting lists, partly due to inefficient patient discharges. This project modelled patient flow in and out of acute hospitals, focusing on discharge delays.

Project description

Hospitals across the country are struggling to provide patients with the care they need in a timely manner. Increasing waiting times in A&E departments, long waits for ambulances and growing waiting lists are all examples of these issues. A possible cause of these issues is the difficulty in discharging patients from acute hospital beds efficiently, due to lack of onward care capacity. Timely discharges are crucial to patient care, to get the right care in the right place at the right time.

A previous PenCHORD project, IPACS, looked at understanding the capacity required in social and community care to facilitate timely flow out of acute hospitals. This project was based on IPACS, and extended the scope to include elective inpatient waiting lists and patients in the emergency department.

The aim of the project was to model flow of patients into and out of an acute hospital site, with particular focus on how delays to complex discharges affect patients waiting to come into the hospital. The results from this project will help to articulate how the issues in discharging patients from an acute hospital bed affect emergency care performance (time spent in A&E and ambulance handover time) as well as waiting lists for elective inpatient care. A discrete event simulation was created to model patient flow through an acute facility and into community/social care.

10-year plan Alignment

SHIFT - Hospital to Community: modelling how delays in complex discharge from acute hospital beds affect flow, highlighting the need for adequate social and community care capacity to support timely patient flow out of hospital.

H5-5005 — Discrete Event Simulation to model elective surgery pathways

Project ID: H5-5005
Cohort: HSMA 5
Authors: Dominic Allen (Guy’s and St Thomas’ NHS Foundation Trust)
Organisations: Guy’s and St Thomas’ NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Elective Surgery, Waiting Times, Surgical, Hospitals
Public code repository: Yes
Source file: previous_projects/hsma_5/h5_5005_des_elective_surgery_pathways/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=prKoVH2DeyU&t=3s
- Code: https://github.com/gasman-dom/gstt_hands_waitlist_st
- Website: https://gstt-hands-waitlist.streamlit.app/

Abstract

The project created a Discrete Event Simulation to model elective surgery pathways. This tool optimises surgical pathways and provides a ready-made format for creating interactive webapps, benefiting future HSMA participants and other interested users.

Project description

The aim of the project was to create a Discrete Event Simulation in SimPy to model elective surgery pathways, which could be used to model changes to a pathway and their impact on waiting times. This was accomplished and realised as an interactive webapp using Streamlit.

The project is FOSS and is available for anyone to use in the future to model their own surgical pathways or other waiting lists. Future HSMA participants, and anyone else who is interested, will be able to modify this model into a format useful to their organisation.

This will provide possible opportunities for the optimisation of surgical pathways, and also provide a ready-made format for future HSMA participants to create interactive SimPy-based webapps without having to “reinvent the wheel.”

H5-5006 — Using Machine Learning to estimate inequities in access to hospital procedures

Project ID: H5-5006
Cohort: HSMA 5
Authors: Benjamin Mouncer (NHS, Norfolk County Council)
Organisations: Norfolk County Council
Techniques: Machine Learning
Application areas: Inequalities, Planned Admissions, APC Dataset, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_5/h5_5006_ml_inequities_hospital_procedure_access/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=NXGYKKPntO8&list=PLgHO2TgIJXdl1MvfkwCGJ6n8_1EYCSzex&index=4

Abstract

The project aimed to determine disparities in accessing planned hospital appointments by calculating true admission rates for high deprivation areas. They found that minimal public data and simple models were insufficient. Future work includes creating synthetic data and exploring additional features like emergency admissions.

Project description

The aim of this project was to determine on an ICB, LTLA and UTLA the overall level of disparity in accessing planned appointments in hospital. The team did this for each high deprivation LSOA calculating the true rate of planned hospital admissions.

They learnt that estimating equity in planned admissions may be possible but not with a minimal set of public data and simple machine learning models.

They put these findings into practice by building in a way to quantify the confidence interval of a machine learning model is essential to effectively use them in real world environments. They also found combining multiple years of data can work with low quality datasets. Of the datasets used the age distribution of an area was the most consistently predictive feature encountered.

One avenue for progression with this project is the creation of synthetic data to have each person within an area as a row of data. Model complexity must increase to create viable models if they even exist. Other feature sets could be explored such as the emergency admissions, type of planned admissions and other variable synthesised from the HES APC dataset.

10-year plan Alignment

SHIFT - Sickness to Prevention: quantifying disparities in access to planned hospital procedures across deprivation areas, supporting the plan’s aim to halve the gap in healthy life expectancy between richest and poorest regions.

H5-5008 — Modelling the location of neonatal critical care units in North West England

Project ID: H5-5008
Cohort: HSMA 5
Authors: Duncan Galletly (North West Neonatal operational Delivery Network (NWNODN))
Organisations: North West Neonatal operational Delivery Network (NWNODN)
Techniques: Mapping, Location Optimization, Discrete Event Simulation (DES)
Application areas: Neonatal, Inpatients, Hospitals
Public code repository: No
Source file: previous_projects/hsma_5/h5_5008_multiobjective_neonatal_optimisation/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=Fsn_PTGCERI&list=PLgHO2TgIJXdl1MvfkwCGJ6n8_1EYCSzex&index=8

Abstract

The project explored the optimal location for 22 neonatal care sites in North West England, using an algorithm that balanced travel time, distance, NICU care episodes, and admission numbers. A discrete event simulation model and dashboard were created to explore various scenarios, aiding stakeholder decision-making.

Project description

In this project, the optimum location for 22 neonatal care sites across North West England was explored.

The algorithm chosen gave equal weighting to the goals of

minimising average travel time
maximising the proportion of people within 30 minutes
minimising the maximum possible distance travelled
maximising the number of care episodes taking place in level 3 NICU units
maximising the minimum number of admissions per year (i.e. trying to avoid units with very low numbers of admissions)
minimising the largest number of admissions per year (i.e. trying to avoid units with very high numbers of admissions)
maximising the proportion within 30 mintues and in level 3 NICU units

This was then built into a discrete event simulation model and a dashboard, allowing a range of scenarios to be explored by stakeholders.

H5-5009 — Network Analysis of diagnostic procedures in A&E setting

Project ID: H5-5009
Cohort: HSMA 5
Authors: Hamzah Shami (NHS England)
Organisations: NHS England
Techniques: Network Analysis, Plotly Dash
Application areas: Emergency Departments, Diagnostic Procedures, ECDS Dataset
Public code repository: Yes
Source file: previous_projects/hsma_5/h5_5009_network_analysis_diagnostics_ae/index.qmd
Links:
- Code: https://github.com/HamzahJShami/HSMA5

Abstract

The project aimed to analyse the relationship between different diagnostic procedures in A&E using NHS ECDS data. A web tool was developed to provide insights into diagnostic procedure usage across the country. The tool will allow users to explore graphic visualisations and accompanying analytics.

Project description

In A&E a patient can be given one or more diagnostic procedures. These can be as simple as the various blood test or more complicated procedures like an MRI test. The aim of this project was to look at the relationship between the different investigations. Hamzah pulled data from the NHS ECDS and created a series of graphs based by A&E provider.

The aim of the project become to create a web tool that can provide insight in how diagnostic procedure are used in A&E departments across the country. Hamzah learnt a lot about graph analysis and the connection between clinical theory and analytical process.

Hamzah is now working on creating a web tool that allows users to explore the graphic visualisation as well as some analytics that accompany the graphs.

H5-5010 — Creating a tool to automatically generate health equity audits for Community Diagnostic Centres

Project ID: H5-5010
Cohort: HSMA 5
Authors: Sarah Houston (UCL Partners Health Innovation); Deborah Newton (Reading Borough Council)
Organisations: UCL Partners Health Innovation, Reading Borough Council
Techniques: Streamlit, Automation, Reproducible Analytical Pipelines (RAP)
Application areas: Health Equity Audits, Community Diagnostic Centres (CDCs), Inequalities, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Analogue to Digital
Public code repository: Yes
Source file: previous_projects/hsma_5/h5_5010_health_equity_audits_cdc/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=tavZS5MmwOA
- Code: https://github.com/SarahHoustonGH/Diagnostic_HealthEquityAudit

Abstract

The project aimed to create a tool to perform a health equity audit for Community Diagnostic Centres (CDCs) in England. The tool will identify healthcare inequalities, suggests data improvements, and supports local CDCs in understanding their impact. Developed using Python and Streamlit, it allows for easier comparison and sharing of learning across regions, with potential applications beyond CDCs.

Project description

Community diagnostic centres (CDCs) have been launched across England to tackle the diagnostic backlog and address healthcare inequalities. CDCs and commissioners struggle to identify their baseline of healthcare inequalities and monitor their impact going forwards. This leads to resourcing constraints on teams in the short term and risks minimising the CDC’s impact on healthcare inequalities in the long term. As these are emerging services, the quality of data collected by them to understand their impact is unclear. This impact can be estimated through a health equity audit, however this is usually and long and manual process.

The aim of the project was to create an open source and shareable tool to:

Perform a health equity audit to identify where the service may be having an impact on healthcare inequalities
Identify where data may be incomplete or poor quality, and automatically suggest evidence-based strategies to improve the data

This has been achieved through development of Python code to process data and a Streamlit app to present data. This project involved developing a toolkit to explore inequalities for multiple sites. Through trialling this toolkit with dummy data for different sites we have developed metrics would be most suitable to identify different regions.

The team would like to further develop the tool to make it more widely applicable and include more robust analysis of inequalities. They are planning to engage with local and national stakeholders of CDCs to demonstrate the tool and gather feedback. They are also planning to implement changes suggested by the Patient and Public Involvement Group (PenPEG).

The tool itself, at a minimum, will support a local CDC to understand their local impact on healthcare inequalities and address areas of concern where relevant. It’s an example to demonstrate the power of open-source techniques and data in healthcare inequalities. If adopted by multiple regions, it would allow different regions to generate similar outputs which would allow for easier comparison and sharing of learning. The code developed could also be easily adapted and reused for other healthcare services beyond CDCs.

10-year plan Alignment

SHIFT - Sickness to Prevention: automating health equity audits for Community Diagnostic Centres to identify and address healthcare inequalities, supporting the plan’s ambition to narrow the gap in healthy life expectancy between richest and poorest areas.

SHIFT - Analogue to Digital: replacing a long, manual audit process with an automated, open-source Streamlit tool, making equity monitoring faster and more consistently shareable across regions.

H5-5012 — Forecasting Demand – Investigating approaches to forecast clock starts

Project ID: H5-5012
Cohort: HSMA 5
Authors: Jane Kirkpatrick (NHS England); Mathew Ojo (NHS England); Lyndsey Allen (NHS North of England CSU (NECS)); Luke Asante (NHS North of England CSU (NECS)); Evelyn Koon (NHS England)
Organisations: NHS England, NHS North of England CSU (NECS)
Techniques: Forecasting, Prophet, ARIMA
Application areas: Clock Starts, Demand & Capacity, Waiting Lists, NHS 10-year plan shifts: Analogue to Digital
Public code repository: No
Source file: previous_projects/hsma_5/h5_5012_forecasting_demand_clock_starts/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=Obzos_OfyEw

Abstract

The project aimed to explore and evaluate different forecasting methods for NHS England’s clock starts, moving beyond Excel-based scenario modelling. The team tested various models and suggested translating the Excel model into Python and obtaining more data to improve demand forecasting and manage wait lists.

Project description

Within the organisation (NHS England) forecasting for clock starts is currently done using scenario-based modelling in Excel. This limits how much data can be used to make forecasts, and the techniques that can be deployed. There was a desire to explore other forecasting methods to see if they could obtain more accurate forecasts, and also to valid the forecasts that are being made using the existing model. Understanding demand is important to manage wait lists, which is a high priority for the NHS, hence the interest in modelling clock starts.

The aim of the project was to explore different ways of forecasting demand, evaluate their performance and offer suggestions as to how forecasting could be done in the future.

The team carried out exploratory data analysis to understand differences between different data types and the differences in the patterns of demand for different subgroupings. They built a codebase that loaded the required data and ran model functions for Naïve, ARIMA, Prophet and a combination of Linear Regression and Random Forest. They evaluated the performance of these models to each other, and the modelling technique currently used. They also made some progress on rebuilding the current excel model in Python.

They found that while the more performant models that we used performed well (produced low error metrics), they generally didn’t perform as well as the model currently used within the organisation. This suggested that a helpful next step of the project might be to complete the work of translating the excel model into python and doing further work to understand if they can get more data to better understand the drivers of demand beyond just time series.

10-year plan Alignment

SHIFT - Analogue to Digital: moving demand forecasting for waiting list management away from Excel-based scenario modelling toward Python-based statistical and machine learning forecasting methods.

H5-5013 — Understanding Excess Mortality in Dorset

Project ID: H5-5013
Cohort: HSMA 5
Authors: Eleanor Jeram (Public Health Dorset); Lee Robertson (Public Health Dorset); Wilson Otitonaiye (Public Health Dorset)
Organisations: Public Health Dorset
Techniques: Streamlit, Forecasting
Application areas: PCMD Dataset, Public Health, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_5/h5_5013_understanding_excess_mortality/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=e4wYE9tphVk

Abstract

The project aimed to improve access and service experience through Dorset ICS’s Health Inequalities programme. It focused on defining and measuring excess mortality, identifying unexpected mortality trends, and understanding driving factors.

Project description

The project was based on the Dorset ICS (Integrated Care Systems) Health Inequalities programme aim to improve access, enhance the experience of services for everyone. Through this and the ICP (Integrated Care Partnership) integrated care strategy supporting early intervention and prevention approaches, reducing the variation in how well people are supported with long-term conditions. This in turn reduces excess mortality.

The aim of the project was to understand and build on the definition of excess mortality and how the baseline is measured. Identifying periods of unexpected mortality (increases and decreases) to understand factors which may be driving changes.

The team learnt that the discrepancies of measuring expected mortality through five year rolling averages removes the impact of seasonality on mortality resulting in hidden trends; with seasonality patterns varying across different underlying causes. The output of the project will allow for differing definitions of expected mortality, providing a base line and forecast moving forward. In addition, different cohorts of the population are considered in line with the Public Health objectives and priorities.

The team are hoping to further develop their StreamLit app to make sharing and access available to the wider team – building questions and bringing in wider subject matter expertise.

Tip

“HSMA was an invaluable opportunity, I was impressed by how much the three members of the team learnt and the development of advanced analytics and forecasting. The hands-on approach in the course has resulted in instant results in our understanding of excess mortality” - Natasha Morris, Public Health Intelligence Team Leader

10-year plan Alignment

SHIFT - Sickness to Prevention: identifying unexpected mortality trends and their drivers to support early intervention and prevention approaches, in line with Dorset’s Health Inequalities programme and the plan’s wider prevention agenda.

H5-5014 — Investigating factors impacting NHS workforce retention

Project ID: H5-5014
Cohort: HSMA 5
Authors: Marie Rogers (NHS England); Richard Pilbery (Yorkshire Ambulance Service NHS Trust)
Organisations: NHS England, Yorkshire Ambulance Service NHS Trust
Techniques: Plotly Dash, Regression, Machine Learning
Application areas: Staff Turnover, Workforce
Public code repository: Yes
Source file: previous_projects/hsma_5/h5_5014_workforce_turnover_drivers/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=yYnkdsM5HHo&list=PLgHO2TgIJXdl1MvfkwCGJ6n8_1EYCSzex&index=2
- Code: https://github.com/marierrogers/HSMAproject2023

Abstract

Project description

This project aimed to work out which factors are the biggest drivers of staff turnover using regression modelling on staff workforce figures as well as other local factors such as employment. This was turned into a dashboard for internal use. However, it was concluded that the currently available data only acccounted for 13% of the variation in turnover seen; more data on other factors is required to explain the patterns seen.

H5-5015 — A Discrete Event Simulation Model to reduce Rheumatology waiting times in Dorset

Project ID: H5-5015
Cohort: HSMA 5
Authors: Abby Dewhurst (Dorset Intelligence and Insight Service (NHS Dorset)); Claire Davies (BCP Council); Krzysztof Cepa (Dorset Intelligence and Insight Service (NHS Dorset))
Organisations: Dorset Intelligence and Insight Service (NHS Dorset), BCP Council
Techniques: Discrete Event Simulation (DES)
Application areas: Rheumatology, Waiting Lists
Public code repository: No
Source file: previous_projects/hsma_5/h5_5015_des_dorset_rheumatology/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=P6nQle5tgBM&t=1s

Abstract

The project aimed to reduce the Rheumatology waiting list and times in Dorset using Discrete Event Simulation (DES). The team built a DES model and a Streamlit app to simulate capacity changes.

Project description

The current Referral to Treatment waiting list is at its highest level in NHS history. Identifying methods for reducing the waiting list and waiting times would benefit patient care.

The aim of the project was to look at what changes in capacity, the number of appointments, could be made to reduce the waiting list and waiting times for Rheumatology services in Dorset. This was done using Discrete Event Simulation (DES), which is a technique that models flow through a pathway and can show what happens if changes are made.

During the project the team built a DES model to look at the Rheumatology pathway. They then built an app using Steamlit so that stakeholders can model the impact of capacity changes on the size of the waiting lists and the average waiting time for patients.

They will be presenting the final model and app to stakeholders and getting feedback, adapting if required; then putting it on GitHub to be shared.

H5-5016 — Developing a tool to assess inequalities and demographic coverage of service locations

Project ID: H5-5016
Cohort: HSMA 5
Authors: Alex Owens (NHS Arden & GEM CSU); Phoebe Woodhead (Health Innovation Wessex); Sian Heath (Nottingham University Hospitals NHS Trust)
Organisations: NHS Arden & GEM CSU, Health Innovation Wessex, Nottingham University Hospitals NHS Trust
Techniques: Streamlit, Mapping, Travel Times
Application areas: Inequalities, Demographics, NHS 10-year plan shifts: Sickness to Prevention
Public code repository: No
Source file: previous_projects/hsma_5/h5_5016_inequalities_demographic/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=7a1bx1NfyBA&t=3s

Abstract

The project aimed to empower NHS service providers with a tool to understand their patient population and direct service expansion to underserved groups. It combines geospatial analysis, demographic data, and travel calculations.

Project description

This project hopes to empower NHS service providers with a tool that grants a deeper understanding of the population they serve and allows them to direct future service expansion to provide access to underserved groups.

It will provide a nuanced demographic breakdown of the patient population, including age, ethnicity, and deprivation, offering crucial insights into the diversity of healthcare needs within the service area – with potential additions to allow providers to add new demographic information to better understand catchment by the conditions they look to treat.

Undertaking this project gave us a deeper understanding of health modelling methodologies and their practical applications. The team learned to combine geospatial analysis, demographic data, and travel distance calculations to create a dynamic tool for use by NHS service providers.

They will continue developing this tool, adding data sources that collaborating service providers hold, but are unable to publish alongside the open-source code, to better identify underserved groups with specific ailments to be treated. This will allow them to give more precise information to let their new service allocation tool pick more appropriate potential locations for new sites. They will be presenting the final model and app to stakeholders and getting feedback, adapting if required; then putting it on GitHub to be shared.

10-year plan Alignment

SHIFT - Sickness to Prevention: combining geospatial, demographic and travel data to direct service expansion toward underserved groups, supporting the plan’s aim to narrow health inequalities and improve access for disadvantaged populations.

H6-6001 — Discrete Event Simulation modelling of Non-elective flow

Project ID: H6-6001
Cohort: HSMA 6
Authors: Helena Robinson (Countess of Chester Hospital NHS Foundation Trust)
Organisations: Countess of Chester Hospital NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Patient Flow, Non-elective Admissions, Emergency Departments, Same-day Emergency Care
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6001_DES_modelling_non_elective_flow/index.qmd
Links:
- Video: https://youtu.be/0DUx5hhxjoA?feature=shared
- Code: https://github.com/helenajr/Non-Elective-Flow-Simulation
- App: https://non-elective-flow-simulation-coch.streamlit.app/
- Slides: https://drive.google.com/file/d/1BIUDaDfDp2cj6EF19908XQ4eB7h9GMCp/view?usp=drive_link

Abstract

Poor patient flow in Emergency Departments leads to long admission waits and poorer patient outcomes. Strategies include increasing beds and reducing discharge delays. This project uses Discrete Event Simulation to explore bed numbers, length of stay, and Same Day Emergency Care impacts on ED waits, aiming to optimise patient flow.

Project description

Poor patient flow is leading to long waits for admission in Emergency Departments. This means there is poor performance against all the key ED wait metrics for the hospital and more importantly, there is evidence that long waits for admission in ED are associated with poorer outcomes for patients.

The two main strategies employed to tackle this problem is

increasing the number of beds (by creation of escalation beds)
trying to decrease discharge delays (reducing length of stay).

Additionally, with the Same Day Emergency Care (SDEC) facility, it is unclear how the number of people admitted through this facility impacts the waits of those in ED.

This projects aimed to answer questions such as:

Given x beds, how far does admitted length of stay have to reduce to meet particular waiting time targets for those queuing in ED? (Evidence based target)
If we open 15 beds but keep admitted length of stay the same, what is the impact on ED waiting times and the various targets? (Evidence for a particular management strategy)
What is the optimum number of people to stream from ED to SDEC to minimise ED waits? (Evidence for a particular management strategy)

This project created Discrete Event Simulation model(s), using HSMA training, to provide evidence for the questions above.

The team also created a friendly user interface that stakeholders can try out scenarios and help understand how the model works.

Talks

App

Click here to open the app in a new window.

H6-6002 — Geographic and Boosted Tree Modelling of Healthcare worker vaccination uptake

Project ID: H6-6002
Cohort: HSMA 6
Authors: Yasmin Ibrahim (NHS England)
Organisations: NHS England
Techniques: Geographic Modelling, Machine Learning
Application areas: Vaccination, COVID-19, Workforce, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6002_GM_healthcare_worker_vaccination_uptake/index.qmd

Abstract

There is a decreasing uptake of COVID and flu vaccinations among healthcare workers. Identifying patterns by staff uptake, gender, ethnicity, deprivation, and other factors can help. Using geographic and boosted tree modelling, along with regression analysis, can capture these patterns and provide useful data for stakeholders to address the issue.

Project description

There is low uptake of Healthcare Workers for COVID and Flu, which is in many Trusts decreasing with each season.

Trusts and regions, as well as NHS England policy teams would benefit from identifying patterns among healthcare workers: staff uptake, gender, ethnicity, deprivation, frontline status, age group, eligibility (in another cohort other than healthcare worker status).

Using a mixed-method approach. Geographic modelling of healthcare worker uptake by Trust, ICB and region, and boosted tree modelling of the data to capture patterns in the data and reduces bias and variance. Regression modelling would also be useful to produce data tables to present to stakeholders (e.g. odds ratios for logistic regression).

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: modelling patterns in healthcare worker COVID and flu vaccination uptake by deprivation, ethnicity and other factors to support targeted action and improve preventative vaccination coverage.

H6-6003 — DESmond: Discrete Event Simulation and Artificially Intelligent Forecasting: Modelling a Prostate Cancer Pathway

Project ID: H6-6003
Cohort: HSMA 6
Authors: Jake Wilson (Great Western Hospitals NHS Foundation Trust)
Organisations: Great Western Hospitals NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Cancer, Waiting Times, NHS 10-year big bet: Data to Deliver Impact, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6003_DESmond_62_day_prostate_cancer_pathway/index.qmd

Abstract

This project aims to develop a Discrete Event Simulation model to predict demand and identify potential bottlenecks using live data. This model could help reallocate resources proactively, helping to ensure people have the fastest possible treatment.

Project description

Prostate cancer pathways involves numerous activities such as triaging, imaging, biopsies as well as multiple decisions along the way. The smooth running of the pathway relies upon having the necessary resources allocated each of these events e.g. biopsy clinics, access to imaging etc.

We want to develop a system which can anticipate the demand for resources, dependant on the number of referrals and resources allocated.

The aim is

to create a Discrete Event Simulation model to model the current pathway
to feed live data into the model in order to predict where bottlenecks might occur
make use of forecasting methods to predicts what changes are needed in resource allocation

The project outcome will be a simulation model with an interactive, web app based interface and a dashboard which illustrates key metrics.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: modelling the prostate cancer pathway to proactively reallocate resources and reduce bottlenecks, supporting faster diagnosis and treatment.

“NHS 10-year big bet: Data to Deliver Impact”: feeding live data into a simulation model to predict demand and bottlenecks in real time, using high-quality data as the basis for proactive resource decisions.

H6-6005 — Improving ambulance care via fast feedback from Quality Care Indicators

Project ID: H6-6005
Cohort: HSMA 6
Authors: Phil King (South Central Ambulance Service NHS Foundation Trust (SCAS)); James Wise (South Central Ambulance Service NHS Foundation Trust (SCAS))
Organisations: South Central Ambulance Service NHS Foundation Trust (SCAS)
Techniques: Machine Learning, Natural Language Processing (NLP), Sentiment Analysis
Application areas: Ambulance
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6005_Improving_ambulance_care/index.qmd
Links:
- Video: https://youtu.be/zKZv3zWcXzc?feature=shared
- Slides: https://docs.google.com/presentation/d/1q1rPYhehziILwtqK9yrtFaU6JklD2c3M/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project aims to improve ACQI data quality and provide rapid feedback using a tool to analyse free text fields, capture sentiment, categorize incidents, and assess treatment appropriateness. This tool could be shared with other Ambulance Trusts and serve as a backup for to manually review data. It will also predict rule changes and their impact on scores.

Project description

Our aim is to improve Ambulance Clinical Quality Indicators (ACQI) data quality and rapidly provide feedback from all ACQI records. There is potential for this tool to be shared with other Ambulance Trusts using the same or similar data collection techniques. This would be achieved with a tool that can analyse free text fields to complete missing data, capture the sentiment of clinical notes, categorise the type of incident and assess if the treatment given was appropriate. It could be utilised as a backup process to manually reviewing data should there be a resourcing issue within the team. It will also evidence if any changes could be made to the clinical records to improve data capture.

We would like to consider how the free text could be searched to find positive and negative statements that impact if a record passes or fails a measure within each ACQI. In addition, creating a tool that would predict any changes to the ‘rules’ and their impact on the scores.

10-year plan Alignment

NHS 10-year big bet: AI to drive productivity: Classification aims to reduce the time it takes to complete audits, as well as supporting feedback and training to enhance ambulance care

NHS 10-year big bet: Data to Deliver Impact: Augmenting existing datasets with new usable flags

H6-6006 — Discrete Event Simulation modelling of childrens ADHD diagnosis and treatment

Project ID: H6-6006
Cohort: HSMA 6
Authors: Heath McDonald (Lancashire and South Cumbria NHS Foundation Trust)
Organisations: Lancashire and South Cumbria NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Neurodiversity, Paediatric, Waiting Lists, Waiting Times, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6006_DES_childrens_ADHD_diagnosis_and_treatment/index.qmd
Links:
- Video: https://www.youtube.com/watch?v=EmfpThPdXKo
- Code: https://github.com/MightyAtom220474/ADHD-Pathway-DES
- Slides: https://drive.google.com/file/d/15AsF2U4BnsXbbGQRxIcFkDW130j4JiaF/view?usp=drive_link

Abstract

The project uses Discrete Event Simulation to model the children’s ADHD diagnostic and treatment pathway, aiming to reduce waiting times and lists. It proposes a new pathway with preliminary diagnostic testing to ensure accurate ADHD assessments. The model will evaluate the impact of these changes and may extend to include 1:1 and group session appointments.

Project description

This project is using Discrete Event Simulation to model the children’s ADHD diagnostic and treatment pathway. The model will be used to identify delays and potential strategies to reduce waiting lists and waiting times for children accessing treatment.

The current assessment process is lengthy, proposing a new pathway that would include undertaking some of the diagnostic testing before undertaking a more comprehensive testing so that there is more confidence that patients who have assessments have ADHD.

The model will capture the changes to the pathway to assess the potential impact of these changes on waiting times and will also capture formal diagnosis and medication titration aspects of the pathway. The model may be extended to look at further 1:1 and group session appointments.

10-year plan Alignment

H6-6007 — Modelling eye injection pathways

Project ID: H6-6007
Cohort: HSMA 6
Authors: Luke Herbert (Surrey and Sussex Healthcare NHS Trust)
Organisations: Surrey and Sussex Healthcare NHS Trust
Techniques: Agent Based Simulation (ABS), Discrete Event Simulation (DES), Streamlit
Application areas: Opthalmology
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6007_Modelling_eye_injection_pathways/index.qmd
Links:
- Video: https://youtu.be/0OKev_kbQEc?feature=shared
- Code: https://github.com/lh/vegf-1

Abstract

The project aims to develop a flexible simulation model to optimise anti-VEGF treatment strategies in ophthalmology. It will use dual modelling frameworks, a modular design, and an interactive dashboard. The model will analyse clinical effectiveness, costs, and resource requirements, adapting to new treatments. The objective is to provide a tool to improve patient outcomes and optimize resource use and costs.

Project description

This project aims to develop a comprehensive and flexible simulation model to optimise anti-VEGF treatment strategies in ophthalmology. The model will address the complex challenges in anti-VEGF treatments, including frequent injections, multiple treatment options, and diverse strategies, whilst considering the significant resource consumption and high costs associated with these treatments.

Key features of the project include:

Dual modelling approach using Mesa and SimPy frameworks to determine the most suitable platform.
Modular design to accommodate various treatment strategies, drug efficacies, and clinic capacities.
User-friendly interface with interactive dashboards for stakeholders to explore different scenarios.
Analysis of clinical effectiveness, costs, and resource requirements for various treatment approaches.
Adaptability to incorporate new drugs and treatment paradigms as they emerge.

Project Objectives: The project aims to provide a tool for data-driven decision-making in ophthalmology clinic management, potentially improving patient outcomes whilst optimising resource use and costs.

H6-6008 — Modelling GP Phone calls

Project ID: H6-6008
Cohort: HSMA 6
Authors: Sarah Blincko (NHS England)
Organisations: NHS England
Techniques: Discrete Event Simulation (DES)
Application areas: 111 Service, Emergency Departments, Primary Care (GP), NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6008_Modelling_GP_phone_calls/index.qmd
Links:
- Video: https://youtu.be/zeYG0EJ83bA?feature=share
- Slides: https://docs.google.com/presentation/d/1RlFY3GLDJg4r89WZ1QNGWaeBT5Fa-sp2/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project uses Discrete Event Simulation to model how patients navigate between GP and 111 services, and to assess the knock-on effect that this has on Emergency Departments. The aim is to support better routing of patients to primary care.

Project description

This project will use Discrete Event Simulation to model patients’ navigation of GP vs 111 services and assess the knock-on effect for Emergency Departments.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: modelling how patients navigate between GP and 111 services to understand the knock-on effect on Emergency Departments, supporting better routing of patients to primary care.

H6-6009 — Using machine learning models to predict future frailty

Project ID: H6-6009
Cohort: HSMA 6
Authors: Emil Frances-Chi (NHS West Yorkshire ICB); Sid Kumar (NHS West Yorkshire ICB)
Organisations: NHS West Yorkshire ICB
Techniques: Machine Learning
Application areas: Frailty, Older Adults, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6009_ML_to_predict_future_fraility/index.qmd

Abstract

The project uses machine learning to estimate future frailty in the Wakefield District and identify key predictive features. It aims to plan resource allocation based on evidence, using two years of linked data. The project will also explore predicting other long-term conditions and produce reports and a user interface for stakeholders.

Project description

This project will use machine learning methods to estimate the size of the Wakefield District population who will likely be frail in 2 / 3 / 4 years time, and to understand the key features used to predict whether a person will become frail in the near future at a local level.

Planning for future population health needs

The ability to plan and design resource allocation (e.g. beds available) in an evidence-based manner is a crucial and consistent priority within the ICB
There is soon to be approximately 2 years’ worth of linked data, allowing us to begin building and evaluating the use of ML-based techniques aiming to predict the future prevalence of various LTCs or conditions
Pending the outcomes of this project, we anticipate there to be future scope to adapt similar approached and apply learning to predicting the future prevalence of other LTCs such as diabetes, hypertension, CVD conditions, respiratory conditions and so forth

The project will produce reports from the work and potentially a dashboard or other user interface.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: estimating the future prevalence of frailty at a local population level to support evidence-based, preemptive resource planning before need arises.

H6-6010 — Understanding drivers of increased length of stay

Project ID: H6-6010
Cohort: HSMA 6
Authors: Jake Kealey (NHS North Central London ICB)
Organisations: NHS North Central London ICB
Techniques: Causal Analysis, Machine Learning, Explainable AI, Synthetic Data, Streamlit, System Dynamics
Application areas: Length of Stay, Inpatients, Understanding Drivers
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6010_Understanding_drivers_of_increased_length_of_stay/index.qmd
Links:
- Video: https://youtu.be/SH4geMXr8a8?feature=shared

Abstract

NCL has seen a rise in long Length of Stay (LoS) over the past 5 years, causing system strain. This project aims to develop a causal model to identify factors affecting LoS and estimate the impact of interventions. Key aims include building qualitative and quantitative models of LoS and modelling changes in response to interventions.

Project description

North Central London (NCL) has observed a consistent increase in long Length of Stay (LoS) over the past 5 years. Increasing length of stay means a higher capacity is needed for the same demand which causes system strain.

Stakeholders are interested in understanding why length of stay has increased within NCL so that that can introduce appropriate measures to reduce it.

This project proposes to develop a causal model that can be used to identify relations between admission features, operations and length of stay. The model will be used to understand site level drivers to estimate the potential impact of proposed interventions.

Key Aims:

To build a qualitative model of the drivers of Length of Stay in acute providers in NCL
Build a quantitative model(s) of LoS
Model changes in LoS in response to interventions (stretching)

H6-6011 — Forecasting modelling for A&E attendance

Project ID: H6-6011
Cohort: HSMA 6
Authors: Yu Qiao (Liverpool University Hospitals NHS Foundation Trust)
Organisations: Liverpool University Hospitals NHS Foundation Trust
Techniques: Forecasting, Reproducible Analytical Pipelines (RAP)
Application areas: Emergency Departments, Demand & Capacity, Seasonality
Public code repository: No
Source file: previous_projects/hsma_6/H6_6011_Forecasting_modelling_for_A&E_attendance/index.qmd
Links:
- Video: https://youtu.be/uywbayhnass?feature=shared
- Case Study: https://hsma.co.uk/impact_posters/Forecasting%20AE%20attendance.png
- Slides: https://docs.google.com/presentation/d/1XaDSIpQDPCXDF7sBMhYlOwdyPSa66VR7/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project aims to forecast A&E attendances 6 weeks in advance, considering seasonal patterns. This helps manage limited resources, plan staffing levels, and assess the need for escalation during abnormal events. Using Time Series Forecasting and a Repeatable Analytical Pipeline, the project addresses the unpredictability of A&E demand influenced by various factors.

Project description

The aim of this project was to forecast future A&E attendances 6 weeks in advance, accounting for seasonal patterns.

Original project brief

NHS A&E is under pressure nationwide for a long-time and often we do not know what lies ahead of us to manage the limited resources available and understand when interventions and helps are needed. This is in addition to seasonal pressures.

Currently A&E has a challenging time looking ahead and manage staffing level needed to match the demand. Furthermore, this can help staff to assess if an escalation process is needed when the abnormal events occur.

A&E is known for its unpredictability in demand that varies with seasonal trend, location, population, weather and various of other factors.

This project will use Time Series Forecasting methods and a Repeatable Analytical Pipeline for Deployment.

Case Study

Video

H6-6012 — Using classification modelling technques to investigate changes in Healthcare Resources Group (HRG) coding over time

Project ID: H6-6012
Cohort: HSMA 6
Authors: Chris Todd (NHS England)
Organisations: NHS England
Techniques: Machine Learning, Explainable AI
Application areas: Clinical Coding, Costs
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6012_Using_classification_modelling_to investigate_changes_in_ HRG/index.qmd
Links:
- Video: https://youtu.be/zeYG0EJ83bA?feature=shared
- Slides: https://docs.google.com/presentation/d/1fGS1mKo8BrF_I4wy5AAzjDsdETnUaFnE/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project will use classification modelling and explainable AI to identify changes in complexity and comorbidity categorisation over time, ensuring accurate cost pressure insights.

Project description

The Medium-Term Activity Projections (MTAP) model produces a set of baseline activity and price-weighted contact projections to 2040/41 for NHS England for some types of activity. As part of the MTAP project, we want to provide insight into the cost pressure based on this projected activity. Modelling the absolute cost is not possible, so one widely used approach to modelling cost pressure is using cost or price ‘weights’ for distinct and specific activity types. Changes in the ratios of different activity types are captured by the sum of the weighted activity.

Healthcare Resource Group (HRG) codes are used in the NHS to categorise types of inpatient activity by assigning one of around 3,000 5-digit codes. The first 4 digits indicate the primary treatment or reason for care, e.g. primary hip replacement, and the 5th digit capturing complexity and comorbidities through the recording of secondary diagnoses. However, in recent years there has been a huge increase in the recording of secondary diagnoses with the introduction of Electronic Patient Records (EPR), but there is a hypothesis that this change does not represent a real increase in patient co-morbidity, and ‘coding creep’ is present.

To deal with this, National Cost Collection (previously known as Reference Costs) data is used to look at the relationship over time between the proportion of activity in each complexity category (the 5th character of the HRG) and the change in the average relative cost of each HRG. The hypothesis is that if increases in the proportion of more complex patients are real, then this should not lead to a reduction in the average relative cost of those HRG.

This research project proposes to identify whether the key factors influencing the complexity and comorbidity categorisation (CCC) have changed over time by:

Training a classification model to predict CCCs, using secondary diagnoses in the patient records as input features. The model will be trained on data for a given year, e.g. 2016/17, and the model performance for other time periods (e.g. 2022/23) will be compared. A significant change in performance may indicate that the factors influencing CCC have changed. If the 2016/17 performance gradually decreases over time as the use of EPR have been increasing, it could support the coding drift hypothesis.
Using explainable AI methods to identify the key factors driving classification in two time periods to assess if there has been a significant change.
Use classification modelling approach to identify the key factors driving the prediction of HRG CCC and actual costs using National Cost Collection data over a given time period. If the HRG codes are reflective of reality, features should have similar relative importance in each model.

H6-6013 — Modelling bed occupancy on an Acute Ward

Project ID: H6-6013
Cohort: HSMA 6
Authors: Rey Tan (Royal United Hospitals Bath NHS Foundation Trust)
Organisations: Royal United Hospitals Bath NHS Foundation Trust
Techniques: Forecasting, Machine Learning, Discrete Event Simulation (DES), Streamlit
Application areas: Length of Stay, Inpatients, Bed Occupancy, Demand & Capacity
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6013_Modelling_bed_occupancy_acute_ward/index.qmd
Links:
- Video: https://youtu.be/SH4geMXr8a8?feature=shared
- Code: https://github.com/ReyTan8/Project—DES-Vidigi

Abstract

The project aims to develop a tool for time series forecasting of bed occupancy using historical data, incorporating seasonality and growth. A web-based app will simulate acute bed models, including variables like closed beds and additional capacity. Using machine learning and discrete event simulation, the tool will aid decision-making and provide reliable daily forecasts.

Project description

The aim of this project is to develop a tool to perform time series forecasting on bed occupancy based on historical data, incorporating variables like seasonality and growth. To build a web-based application that enables end users to simulate acute bed model, with the ability to include variables like closed beds, additional capacity, community availability etc, in turn aiding decision making.

The problem:

Current bed occupancy forecast is done on Excel based on historical monthly averages
High level position but lacking in flexibility
Daily bed model is manually tweaked, and unable to simulate possible outcome

Using Machine learning and discrete event simulation the aim is for the project to predict values close to actual measures. For the user to utilise the web app on a day to day basis and find it a reliable tool for decision making.

H6-6015 — Forecasting the supply of medical doctors

Project ID: H6-6015
Cohort: HSMA 6
Authors: Sara Fundu (The Royal College of Pathologists)
Organisations: The Royal College of Pathologists
Techniques: Forecasting
Application areas: Workforce
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6015_Forecasting_the_supply_of_medical_doctors/index.qmd

Abstract

The project aims to forecast the regional demand for medical doctors, considering factors like trainees and population. It will plot a graph showing the current supply versus demand. Additionally, it will simulate outcomes under different scenarios, such as changes in trainee numbers or population health, to better understand and address the gap.

Project description

The problem: Not enough medical doctors by region and we need to establish the gap between what we have and what we need.

The Aim: Forecast medical doctors required by region, taking into account relevant factors for example, trainees coming in, population etc.

The Output: To plot a graph to show our current supply and what we demand.

Bonus would be to be able to simulate outcome of demand, if input is amended to reflect different scenarios i.e. number of trainees decreases or population ill health increases etc

H6-6016 — Optimising the location of Breast Cancer diagnostic services across Devon

Project ID: H6-6016
Cohort: HSMA 6
Authors: Gill Baker (Royal Devon University Healthcare NHS Foundation Trust); Brandon Jones (Royal Devon University Healthcare NHS Foundation Trust); Kat Pamatmat (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Geographic Modelling, Travel Times
Application areas: Cancer, Women’s Health
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6016_Optimising_the_location_of_breast_cancer_diagnostic_services/index.qmd

Abstract

Over 6000 patients are referred annually for fast-track breast symptom diagnosis at RD&E and NDDH. Increasing referrals and limited infrastructure necessitate building or extending diagnostic units. The project aims to determine optimal locations for additional services to minimize patient travel time and reduce costs, using data science to map demand and calculate travel distances and benefits.

Project description

Over 6000 patients a year are referred to Royal Devon & Exeter Hospital (RD&E) and North Devon District Hospital (NDDH) for fast-track breast symptom diagnosis. The fast-track clinic requires a clinician, a mammography team and a consultant radiologist/radiographer plus support staff. The imaging needs to take place in specialised rooms. As population and awareness of breast cancer increases there are increasing referrals and the current infrastructure at both RD&E and NDDH will not be sufficient to meet future demand. It is therefore necessary to build a new breast diagnostic unit or extend an existing one. It is likely to be prohibitively expensive to build at both NDDH and RD&E. At present specific GP surgeries refer patients to either NDDH or RD&E. However, the two hospitals are now managed as part of a single NHS Trust called RDUH. Each hospital requires breast diagnosis infrastructure on site as it is used for inpatient and emergency admissions as well as for elective diagnosis services so we need at least two breast diagnosis centres but data science may be used to help determine whether there is a value in building a third centre and/or diverting referrals from specific GPs to one of the existing units.

The aim of this project is to determine where additional breast diagnostic services should be placed to minimise overall travel time for patients whilst reducing infrastructure and running costs.

Objectives include the following:

To map breast referral demand by GP and calculate the weighted average distance and/or travel time from each GP to RD&E, NDDH and a potential central location.
To calculate the percentage of patients who would benefit from a central location.
To calculate the shortest distance/travel time from each GP to the three locations and determine the reduction or increase in mileage if patients were diverted from their current default hospital to one of the other two centres.

H6-6017 — Referral to treatment waiting times for Neurosurgical patients

Project ID: H6-6017
Cohort: HSMA 6
Authors: Andrew Sharrock (The Walton Centre NHS Foundation Trust)
Organisations: The Walton Centre NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Surgical, Neurology, Waiting Lists, Waiting Times
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6017_Referral_to treatment_waiting_times_for_Neurosurgical_patients/index.qmd

Abstract

The project models the neurosurgical patient pathway to predict waiting list changes and treatment wait times. It aims to add user interaction to explore how capacity adjustments affect waiting times and track patients waiting over 52 weeks each month.

Project description

This project will aim to model the pathway for Neurosurgical patients, and to explore, based on current waiting times,

will the waiting list grow or reduce
how long patients may wait for treatment

The aim is to add user interaction, to see how an increase/reduction in capacity affects waiting times. Also to be able to see how many patients are waiting over 52 weeks at each month end.

H6-6018 — Predicting the risk of injurious falls in older people with atrial fibrillation

Project ID: H6-6018
Cohort: HSMA 6
Authors: Anneka Mitchell (University Hospitals Plymouth NHS Trust)
Organisations: University Hospitals Plymouth NHS Trust
Techniques: Machine Learning, Explainable AI
Application areas: Older Adults, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6018_predicting_risk_of_injurious_falls/index.qmd

Abstract

Atrial fibrillation (AF) increases stroke risk, and anticoagulation reduces this risk but can cause bleeding. Despite guidelines, many clinicians avoid prescribing anticoagulants to those at risk of falls. This project explores using machine learning to predict injurious falls in older AF patients and aims to develop a tool to personalize anticoagulant treatment based on falls risk.

Project description

Atrial fibrillation (AF) is a common cardiac arrhythmia which increases the risk of stroke.

Anticoagulation is very effective in reducing stroke risk but can increase the risk of bleeding, a much feared consequence of anticoagulation is bleeding on the brain. National and international guidance states that anticoagulation should not be withheld because of falls as the benefits still outweigh the risks but many clinicians choose not to prescribe these medications to people who fall or those at risk of falls because they don’t believe the evidence supports this recommendation.

The initial stage aims of this project is to explore if machine learning techniques can be used to develop a model that can predict injurious falls in older people with AF and determine what features are important.

The longer term aim of this project is to ascertain whether a tool could be developed to personalise anticoagulant treatment based on falls risk to help clinicians and patients make more informed treatment decisions.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: using machine learning to personalise anticoagulant treatment decisions based on individual falls risk, supporting more preventative, personalised care for older patients.

H6-6019 — Clinical coding automation using Natural Language Processing

Project ID: H6-6019
Cohort: HSMA 6
Authors: Sid Kumar (NHS South West London ICB)
Organisations: NHS South West London ICB
Techniques: Natural Language Processing (NLP)
Application areas: Clinical Coding, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6019_clinical_coding_automation/index.qmd

Abstract

The project aims to use Natural Language Processing (NLP) to automate the prediction of ICD-10 or OPCS-4 codes from doctor/patient notes, currently done manually. Initially focusing on 3-character ICD-10 chapters, it will eventually predict full 4-character codes. Collaborating with a provider, the project will streamline coding and improve accuracy.

Project description

The aim of this project is to use Natural Language Processing (NLP) to predict ICD-10 or OPCS-4 codes from doctor/patient notes, automating a task currently done manually by the clinical coding team.

Initially, we’ll predict 3-character ICD-10 chapters, with the goal of eventually predicting full 4-character codes. Collaborating with a provider that has access to these notes, this project will streamline coding and improve accuracy.

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: using NLP to automate the prediction of ICD-10/OPCS-4 codes from clinical notes, a task currently performed manually, freeing up coding staff time and improving accuracy.

“NHS 10-year big bet: Data to Deliver Impact”: Enhancing held data with categories that can be used downstream for additional analysis or to support data products like machine learning algorithms

H6-6020 — Modelling 111 option 2 call centre

Project ID: H6-6020
Cohort: HSMA 6
Authors: Richard Hall (Norfolk and Suffolk NHS Foundation Trust)
Organisations: Norfolk and Suffolk NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Forecasting, Streamlit, Quarto
Application areas: 111 Service, Telephone-based Services, Staffing Level Optimisation, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6020_modelling_111_option_2_call_centre/index.qmd

Abstract

The project aims to improve two struggling 111 call centres for mental health patients in Norfolk and Suffolk. It will develop a DES model to compare staffing approaches and determine the resources needed for safe service. Additionally, a tool will be created to ensure safe rosters by inputting current rosters and forecasting call levels.

Project description

We operate 2 separate 111 call centres for mental health patients (for both Norfolk and Suffolk). These are both set up in different ways (one fully staffed with ‘qualified’ B6s, the other with junior staff who escalate to qualified staff when needed), but both are struggling with their performance.

Many callers abandon their calls before they can be reached and there are serious concerns about clinical safety. A piece of work is underway to transform these services, and a piece of demand and capacity analysis is required to ensure the best possible staffing model is in place, both in terms of structure and resource levels.

The aim of this project is to build a Discrete Event Simulation model to compare both staffing approaches and identify the level of resources required to safely staff the service. Further aims would be to build a useful helper tool to ensure planned rosters are safe, users could input current rosters again a forecasted level of calls and the model would show the potential effects of this.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: ensuring patients can get timely access

H6-6021 — Predicting the future demand for Renal replacement therapy

Project ID: H6-6021
Cohort: HSMA 6
Authors: Sandhya Radha Krishnakumar (NHS England)
Organisations: NHS England
Techniques: Forecasting
Application areas: Renal, Dialysis, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6021_predicting_future_demand_renal_replacement_therapy/index.qmd

Abstract

Kidney disease is projected to be the fifth leading cause of premature deaths globally by 2040. Rising demand for dialysis and transplants exceeds capacity in England. The project aims to develop a model using forecasting techniques to predict future demand and address it by increasing home dialysis or introducing a new dialysis centre.

Project description

The problem: Kidney disease is projected to be the fifth leading cause of premature deaths globally by 2040. With people living longer and having more comorbidities, the demand for dialysis and transplants is rising and exceeding capacity. In England, the system is at full capacity for in-centre dialysis, requiring immediate action and a long-term strategy to optimise care.

The result: Incorporate forecasting techniques to predict the future demand based on population prediction (data available in ONF and prevalence data available in QOF).

The aim: To develop a model to help tackle this demand - either increasing home dialysis or introducing a new dialysis centre

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: forecasting demand for dialysis to support increasing home dialysis provision as an alternative to in-centre treatment, shifting care away from hospital-based settings.

H6-6023 — Predicting Gestational Diabetes and other maternity-related conditions using machine learning

Project ID: H6-6023
Cohort: HSMA 6
Authors: Rochelle Francis-Reid (Epsom and St Helier University Hospitals NHS Trust)
Organisations: Epsom and St Helier University Hospitals NHS Trust
Techniques: Machine Learning, Streamlit
Application areas: Diabetes, Maternity, Women’s Health, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year big bet: AI to drive productivity
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6023_predicting_gestational_diabetes_using_ML/index.qmd
Links:
- Video: https://youtu.be/zeYG0EJ83bA?feature=shared

Abstract

The project aims to use machine learning to predict gestational diabetes and other maternity-related conditions early in pregnancy, enabling timely interventions and personalised care. It will develop a predictive model, an interactive web app for clinicians to input patient data and receive predictions, and reports on the model’s accuracy and effectiveness.

Project description

Maternity-related conditions, such as gestational diabetes, can have significant impacts on the health of both the mother and baby if not identified early. Current approaches often detect these conditions after symptoms appear, potentially delaying important interventions. This project aims to use machine learning to predict the likelihood of developing gestational diabetes and other related conditions early in pregnancy, allowing for timely interventions and personalised care.

The aim of this project is to develop a machine learning model that predicts the likelihood of gestational diabetes and other maternity-related conditions based on patient data. The model will serve as a decision support tool for clinicians to provide proactive and personalised care, improving health outcomes for mothers and babies.

The project will generate :

A machine learning predictive model.
An interactive web app or interface for clinicians to input patient data and receive predictions.
Reports outlining the model’s predictions, accuracy, and effectiveness.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: predicting the likelihood of gestational diabetes and other maternity conditions early in pregnancy to enable timely, personalised preventative intervention rather than detection after symptoms appear.

“NHS 10-year big bet: AI to drive productivity”: providing clinicians with an interactive tool that uses a predictive model to support proactive, personalised decision-making at the point of care.

H6-6024 — Developing a streamlit app for creating Theographs of patient journeys

Project ID: H6-6024
Cohort: HSMA 6
Authors: Suprasad Gavhane (NHS North of England CSU (NECS))
Organisations: NHS North of England CSU (NECS)
Techniques: Data Visualisation
Application areas: Patient Pathways
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6024_Streamlit_app_for_theographs_of_patient_journeys/index.qmd
Links:
- Code: https://github.com/sp-necs/TheoGraph

Abstract

The project aims to create an open-source application for generating interactive theograph visuals to understand patient/client journeys. It will be a generic tool requiring minimal data fields, usable in various healthcare or social care settings.

Project description

The aim of this project is to create an open source application to generate interactive theograph visuals to understand a patients’/clients’ journey through a system.

The goal is to make a generic application requiring the minimum of data fields so it can be used in different healthcare or social care settings (working with dynamic ‘Event_Type’ values).

More information about theographs can be found here.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: Automated conversion of existing information into a format that is quick, easy and interprtable, helping patterns be identified sooner in a patients’ journey and saving clinician time.

H6-6025 — Modelling delays in breast, head and neck cancer pathways

Project ID: H6-6025
Cohort: HSMA 6
Authors: Michael Baser (NHS England); Laura Webster (NHS England); Lizzie Augarde (NHS England)
Organisations: NHS England
Techniques: Discrete Event Simulation (DES)
Application areas: Cancer, Women’s Health, Waiting Times
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6025_Modelling_delays_in_breast_head_neck_cancer_pathways/index.qmd

Abstract

The project uses DES to model post-diagnosis pathways for breast and head and neck cancer. It aims to identify delays and treatment variations across England, focusing on two key steps - time from neoadjuvant SACT to surgery or radiotherapy, and time from surgery to first adjuvant treatment.

Project description

Breast cancer and head and neck cancer are two common cancer sites, where the main clinical pathway is surgery, followed by adjuvant radiotherapy or chemotherapy. Clinical feedback has reported large waits after surgery and often before the point of referral to e.g. the radiotherapy department.

Our project is using Discrete Event Simulation to model the post-diagnosis pathway for breast and head and neck cancer. The model will be used to identify mid-pathway delays and treatment variation across England with the aim to report on two clinically important steps in the clinical pathway:

time from neoadjuvant SACT to surgery or radiotherapy (RT)
time from surgery to first adjuvant treatment (either RT or SACT)

H6-6026 — Developing a DES Model for a Mental Health Hub

Project ID: H6-6026
Cohort: HSMA 6
Authors: Helen Wharam (Hampshire and Isle of Wight Healthcare NHS Foundation Trust); Nathan Hack (Hampshire and Isle of Wight Healthcare NHS Foundation Trust)
Organisations: Hampshire and Isle of Wight Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: Mental Health, Community Mental Health, Waiting Times, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6026_Modelling_secondary_care_psychological_therapy_flow/index.qmd
Links:
- Video: https://youtu.be/GlEoH8B3CHE?feature=shared
- Slides: https://docs.google.com/presentation/d/1Da6UyQcS1glQF5urB7OzzkKDOhEoLL1n/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

This project models the Portsmouth Mental Health Hub using Discrete Event Simulation to assess call waiting times, durations, and staff utilisation. As demand grows, the model and web-based app will inform resource planning and service improvements, helping ensure timely, effective support for individuals seeking mental health advice and guidance.

Project description

The Mental Health hub in Portsmouth is a relatively new service. It acts as a call centre providing signposting, advice and guidance to those who contact them.

Current service capacity to answer and manage calls was based on estimates of anticipated demand. Routes into the service are now being expanded and demand will potentially grow.

This project will use Discrete Event Simulation to model call waiting times, call duration and resource utilisation.

A web based app will be developed to support ongoing service development.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: modelling call handling capacity for a community-based mental health advice and signposting hub, supporting timely access to support outside of hospital settings.

H6-6034 — Forecasting blood donation session capacity

Project ID: H6-6034
Cohort: HSMA 6
Authors: Sam Plimmer (NHS Blood and Transplant)
Organisations: NHS Blood and Transplant
Techniques: Machine Learning, Streamlit
Application areas: Blood & Transplant, Non-attendance Prediction
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6034_Forecasting_blood_donation_sesion_capacity/index.qmd
Links:
- Video: https://youtu.be/zeYG0EJ83bA?feature=shared

Abstract

Blood donation sessions face issues with cancellations, non-attendance, and medical rejections, leading to missed donations. This project aims to develop a Machine Learning tool to predict actual attendance and assess additional capacity. It will also build a web app for users to review current session capacity and manage bookings effectively.

Project description

Blood donation sessions have limits of how many people will be able to attend a session. Bookings for sessions often don’t convert into actual donations due to cancellations, non-attendance and medical rejections. These missed collections represent a lost opportunity to successfully bring someone in to donate.

The aim of this project is to:

Develop a Machine Learning-based predictive tool to assess, of the current bookings made for a session, what proportion of these are likely to take place, and if there is room for additional capacity
Build a web based app that demonstrates the above and allows users to review current available capacity at each session

H6-6035 — Developing a web app to recommend appropriate technology enabled care

Project ID: H6-6035
Cohort: HSMA 6
Authors: Radia Woodbridge (Torbay and South Devon NHS Foundation Trust)
Organisations: Torbay and South Devon NHS Foundation Trust
Techniques: Streamlit, Python
Application areas: Older Adults, Costs, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6035_web_app_to_recommend_appropriate_tech_enabled_care/index.qmd

Abstract

Technology Enabled Care (TEC) supports independence and health and social care. This project aims to develop a web app providing comprehensive, user-friendly guidance on TEC equipment. It will serve as a one-stop platform for health professionals, carers, and individuals, enhancing personal safety, independence, and reducing the burden on health and social care systems.

Project description

The growing importance of Technology Enabled Care (TEC) in today’s health and social care environment, especially with the increasing aging population and the need for cost-effective solutions to enhance quality of life, has led professionals to incorporate TEC in the individuals’ health and care assessment. TEC devices like personal alarms, sensors, fall detectors, remote health monitoring, and other assistive technologies are used to support individuals, particularly older adults, people with disabilities, or those with health conditions, in maintaining independence and ensuring safety in their daily lives.

Technology enabled care can increase choice in maintaining independence, stay connected with family and friends and enhance social and mental activity.

However, many professionals, carers and individuals are unfamiliar with the wide range of TEC options available, making it difficult to choose the right equipment. In addition, information about TEC equipment is scattered across different platforms making it difficult to access comprehensive, up-to-date guidance in one place.

The aim of this project is to develop a web app that will :

Serve as a one-stop platform providing details, user friendly guidance on TEC equipment.
The app would be useful for Torbay and South devon Trust’s health and social care professionals, carers, and individuals who need support with TEC equipment, creating a bridge between technology and users who want to remain independent and safe in their homes.
The app can provide a platform to keep users updated with the latest innovations.
The app will empower users with comprehensive guides on TEC equipment, enabling them to make confident choices that enhance personal safety and independence
Health and social care professionals will have access to a reliable resource to recommend the right TEC solutions for patients, enhancing their ability to deliver personalised care.
By encouraging the use of TEC, the app contributes to reducing the burden on health and social care systems and cares, enabling more people to remain safely at home.

Future, further extension of this project will include exploring predicting cost savings and change on staff capacity based on TEC usage.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: providing guidance on technology enabled care equipment such as personal alarms and remote monitoring, supporting older adults to remain independent and safe at home rather than relying on hospital or institutional care.

H6-6036 — Forecasting demand in RDUH breast care services and the impact of urban development

Project ID: H6-6036
Cohort: HSMA 6
Authors: Gill Baker (Royal Devon University Healthcare NHS Foundation Trust); Brandon Jones (Royal Devon University Healthcare NHS Foundation Trust); Kat Pamatmat (Royal Devon University Healthcare NHS Foundation Trust)
Organisations: Royal Devon University Healthcare NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Forecasting, Streamlit
Application areas: Cancer, 2-week Wait, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6036_Forcasting_demand_RDUH_breast_care_services/index.qmd

Abstract

Between 2011 and 2021, the population served by Royal Devon & Exeter Hospital grew by 13%, doubling the national average, leading to long waiting lists. This project aims to develop a forecasting tool for breast services to predict referrals and service demand, inform workforce planning, and justify infrastructure expansion. It will also predict geographical demand changes for optimal service locations.

Project description

Historically there has been a lack of data science or accurate forecasting in developing capacity. Between 2011 and 2021 the growth in the population served by the Royal Devon & Exeter Hospital rose by 13% which was twice the average national population growth. This increase in demand was not met by sufficient increase in physical infrastructure or capacity. This led to significant waiting list problems with RD&E having some of the longest waiting patients in the country. Exeter and surrounding areas have high housing targets and predict further growth as well as significant growth in an ageing population.

This project aims

To develop a demonstrator forecasting tool (using breast services as an example) which is made available through a StreamLit app to predict referrals and service demand.
To use this tool to demonstrate the impact of specific housing developments on service demand to inform workforce planning and to provide data to justify s106 payments from developers to support NHS infrastructure expansion.
To predict changes in geographical demand and thus feed into project in understanding optimum locations for services on the 10-20 year timescale.

The project objectives are

To look at the growth in referrals for breast services and predict future clinic capacity required to meet 2WW targets.
To separate the increase in referrals due to population growth from that due to changes in demographics and referral patterns (due to patient education and increasing incidence of cancer).
To create a model of growth based on assumptions about population growth and changes in demographics to obtain best and worst case scenarios in terms of demand.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: predicting future geographical demand for breast services driven by housing growth, to inform optimal service location planning over the next 10-20 years.

H6-6037 — Mapping health inequalities, depreviation, ethnicities and crime across the UK

Project ID: H6-6037
Cohort: HSMA 6
Authors: Abhinav Jindal (NIHR Clinical Research Network)
Organisations: NIHR Clinical Research Network
Techniques: Geographic Modelling, Mapping
Application areas: Inequalities, Crime, Police, Demographics, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6037_mapping_health_inequalities/index.qmd

Abstract

The project overlays ONS data on population ethnicity, the Health Index for England and deprivation data from QOF with crime statistics at a geographical scale. The aim is to test the hypothesis that these factors are linked, and to establish the strength of any relationship.

Project description

The aim of the project is to use ONS data around ethnicities of population, health index England, Deprivation data from QOF, and overlay the same with crime statistics at a geographical scale to try and prove a relationship and the hypothesis of a link.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: mapping deprivation, ethnicity and health index data alongside crime statistics to explore relationships that could inform where to target preventative public health action.

H6-6039 — Applying Natural Language Processing to automate the extraction and classificiation of congenital anomaly diagnoses from free text and genetic data

Project ID: H6-6039
Cohort: HSMA 6
Authors: Charlotte Eversfield (National Disease Registration Service (NDRS)); Jack Anderson (National Disease Registration Service (NDRS)); Claire Welsh (National Disease Registration Service (NDRS)); Clarice Quinn (National Disease Registration Service (NDRS))
Organisations: National Disease Registration Service (NDRS)
Techniques: Natural Language Processing (NLP), Streamlit
Application areas: Congenital Anomilies, Genetic Data, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6039_NLP_to_automate_classification_congenital_anomaly/index.qmd

Abstract

This project aims to use Natural Language Processing to automate and standardise extraction and classification of Congenital anomaly diagnoses under ICD10 code Q87.8 which are manually classified from free text, risking errors and inefficiency. To validate diagnoses with genetic data by defining a data linkage method.

Project description

Congenital anomaly diagnoses can be submitted to the National Congenital Anomaly and Rare Diseases Services (NCARDRS) under an unspecific and uninformative ICD10 code of Q87.8. Q87.8 is broadly defined as ‘Other specified congenital malformation syndromes, not elsewhere classified’, with diagnosis detail submitted in free text data. As a result, classification of these diagnoses into more specific syndromes is currently handled manually by reviewing free text fields from data submissions from hospitals such as discharge letters. This manual review is time consuming and risks human error, duplication of effort, and poor standardisation of rules/methods applied for extraction and classification.

Additionally, diagnoses submitted from discharge letters are only ‘probable’ or ‘suspected’ and therefore need to be validated via genetic testing data. However, there currently is no defined method of data linkage between the discharge letters and genetic data which means diagnoses cannot be confirmed. Providing accurate data on congenital anomaly diagnoses which have been validated by genetic testing data is crucial for disease surveillance reporting.

The aims of this project are:

To use Natural Language Processing (NLP) to automate and standardise the extraction and classification of suspected ICD10 Q87.8 congenital anomaly diagnoses from free text.
To use genetic data to validate these suspected diagnoses, which will include defining a data linkage method between the two datasets.

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: using NLP to automate the manual classification of congenital anomaly diagnoses from free text, reducing time, duplication and error in disease surveillance reporting.

“NHS 10-year big bet: Data to Deliver Impact”: linking genetic testing data with free-text diagnoses to validate congenital anomaly classifications, treating genomic data as the basis for more accurate disease surveillance.

H6-6040 — Modelling the benefit of MECC (Making Every Contact Count) Training using agent based simulation

Project ID: H6-6040
Cohort: HSMA 6
Authors: Dominic Rowney (NHS North of England CSU); Luke Herbert (Surrey and Sussex Healthcare NHS Trust); Sam Vautier (Somerset NHS Foundation Trust)
Organisations: NHS North of England CSU, Surrey and Sussex Healthcare NHS Trust, Somerset NHS Foundation Trust
Techniques: Agent Based Simulation (MECC), Streamlit
Application areas: Making Every Contact Count (MECC), Population Health, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6040_benefit_of_MECC_using_ABS/index.qmd
Links:
- Video: https://youtu.be/U6-_3q_CZtA?feature=shared
- Code: https://github.com/DomRowney/Project_Toy_MECC
- Website: https://domrowney-project-toy-mecc-streamlit-appapp-alcohol-cqzaxu.streamlit.app/
- Slides: https://drive.google.com/file/d/1X_JizOXHMrseAvxSSZWE3YGXuuSo5okM/view?usp=drive_link

Abstract

Making Every Contact Count (MECC) is an e-learning program for health and social care staff to promote healthy lifestyles. This project uses Agent Based Simulation to model MECC’s impact on behaviors like smoking, drinking, and exercise.

Project description

Making Every Contact Count (MECC) is an e-learning programme in which health and social care staff are encouraged to use various interactions with patients to open conversations about healthy lifestyles and wellbeing.

The aim of this project is to use Agent Based Simulation to model the behavioural dynamics and success rates on lifestyle factors (such as smoking, drinking, exercise) of MECC interventions, and thereby try to understand the potential impact of MECC training.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: modelling the potential impact of Making Every Contact Count training on lifestyle behaviours such as smoking, drinking and exercise, supporting preventative conversations between staff and patients.

H6-6041 — Identifying which patients are most at risk for an outcome across integrated neighbourhood teams

Project ID: H6-6041
Cohort: HSMA 6
Authors: Rachel Christie (North West London ICB)
Organisations: North West London ICB
Techniques: Machine Learning, Explainable AI, Streamlit, Geographic Modelling
Application areas: Demographics, Public Health, NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6041_patients_risk_integrated_neighbourhood_teams/index.qmd

Abstract

Population health management uses segmentation to categorise people by health status and needs. However, generic segments may not identify high-risk groups effectively. This project aims to create a tool to identify at-risk patient groups across different geographies, focusing on outcomes like emergency admissions, vaccination rates, and screening uptake.

Project description

Population health management often use population segmentation to categorise the population according to health status, health care needs and priorities. This approach recognises that groups of people share characteristics that influence the way they interact with health and care services.

However, generic population segments may not break down the population into specific enough areas in order to identify which populations are most at risk of an outcome and, therefore, who to target for intervention and where the biggest gains in intervention could be made. Identifying which patient characteristics carry the biggest risk most often comes from clinical expertise and is not data-driven. When data is considered, there is often “too much” data to look into, requiring expertise on where to start. Additionally, different geographies want to know who is most at risk in their patch and often want to be able to view the data at their geography or for their primary care network.

There is a need for a way to identify which patient characteristics carry the biggest risk for an outcome across different geographies.

The aim of this project is to create a tool to identify which patient groups are most at risk for an outcome at different geographies. The outcomes will align in the NWL INT outcomes which include:

Emergency admissions due to fall in over 65s
Not being vaccinated
Not taking up screening
Emergency admissions for ambulatory care sensitive conditions

Machine Learning and Explainable AI approaches will be used to build the tool and outputs.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: identifying patient groups at greatest risk of falls, non-vaccination, non-screening, or emergency admission across neighbourhoods, to better target preventative intervention and resource.

“NHS 10-year plan shifts: Hospital to Community”: building a tool aligned with neighbourhood-level outcomes to support more localised, targeted care planning within integrated neighbourhood teams.

H6-6043 — Predictive modelling for smoking cessation success

Project ID: H6-6043
Cohort: HSMA 6
Authors: Ryan Hutchings (Dorset Council); Khetha Mngomezulu (Dorset Council)
Organisations: Dorset Council
Techniques: Machine Learning, Explainable AI, Causal Analysis
Application areas: Smoking, Public Health, Council, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6043_smoking_cessation_success/index.qmd

Abstract

Smoking cessation remains challenging despite public health efforts. This project aims to develop a predictive model to identify individuals likely to quit smoking based on demographics and behaviours. It will uncover key predictors and effective pathways using machine learning algorithms like logistic regression, decision trees, random forests, and neural networks, evaluating each for accuracy and interpretability.

Project description

Smoking is a leading cause of preventable diseases and deaths worldwide. Despite numerous public health campaigns and interventions, smoking cessation remains a significant challenge. Understanding the factors that contribute to successful smoking cessation can help in designing more effective interventions.

Certain pathways may suit individuals needs differently and thus have different likelihoods of success. This model will help to show which demographics and pathways are most effective at achieving a successful quit.

The primary goal of this project is to develop a predictive model that identifies individuals likely to stop smoking based on their demographics and/or behaviours. Additionally, the project aims to uncover key predictors that can possibly influence smoking cessation, providing insights into the traits and behaviours that contribute to quitting smoking and which pathways offer better results for people.

Several machine learning algorithms will be considered, including logistic regression, decision trees, random forests, and neural networks. Each algorithm will be evaluated for its predictive accuracy and interpretability.

Depending on the complexity of the data to be used, we will also consider employing ensembled methods to leverage the power of different models on one dataset.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: identifying which demographics and treatment pathways are most likely to lead to successful smoking cessation, supporting more effective preventative public health interventions.

H6-6044 — Population segmentation of GP-registered population in Dorset

Project ID: H6-6044
Cohort: HSMA 6
Authors: Rhianna Everett (Dorset Intelligence and Insight Service); James Roberts (Dorset Council)
Organisations: Dorset Council, Dorset Intelligence and Insight Service
Techniques: Machine Learning, Unsupervised Learning
Application areas: Primary Care (GPs), NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6044_Population_segmentation_GP_registered_Dorset/index.qmd
Links:
- Video: https://youtu.be/SH4geMXr8a8?feature=shared
- Slides: https://docs.google.com/presentation/d/1rRlb6pDf89Tph7sNj5r8xN5NdeD84P7d/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

This project aims to develop a machine learning-based population segmentation model for the GP-registered Dorset population, using multiple characteristics including healthcare utilisation. The goal is to identify segments based on care needs to inform service design, enhancing patient understanding and improving resource allocation.

Project description

The aim of this project is to create a machine learning-based population segmentation model for the GP-registered Dorset population using multiple characteristics including primary and secondary healthcare utilisation.

We want to identify population segments based on care needs to inform service design.

This has the potential to provide enhanced patient understanding leading to improved resource allocation.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: segmenting the GP-registered population by care needs to inform proactive service design, supporting earlier identification of population health needs.

H6-6045 — Forecasting NHS planning and performance metrics

Project ID: H6-6045
Cohort: HSMA 6
Authors: Sam Wheeler (NHS Bath and North East Somerset, Swindon and Wiltshire ICB); Sally Cherrington (NHS Bath and North East Somerset, Swindon and Wiltshire ICB)
Organisations: NHS Bath and North East Somerset, Swindon and Wiltshire ICB
Techniques: Forecasting
Application areas: NHS
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6045_Forecasting_NHS_Planning_Performance_metrics/index.qmd

Abstract

This project aims to create a robust forecasting approach for NHS Planning metrics, improving system planning and operational management, and ensuring consistent adoption across the system to aid decision-making.

Project description

The ICB leads on development of Operational Planning submissions to NHS England on an annual basis. This involves development trajectories against various activity and performance metrics, usually covering the next 12 months (e.g. ICB-level A&E attendances). Analysts are often asked to forecast these measures forward based on historic information. The system’s approach to this is often very basic and crude and, partly because of this, organisations often use their own, different approaches which are often open to manipulation.

The aim of this project is to develop an approach to forecasting the main NHS Planning performance and activity metrics, which is more robust than existing, crude techniques and can form the basis of System Planning. To develop an approach that, given it’s more robust, will be consistently adopting across our system to aid decision makers with both planning but also operational management at system level.

H6-6046 — Proactive Patient Attendance Prediction: Enhancing Healthcare Efficiency through Attendance Forecasting

Project ID: H6-6046
Cohort: HSMA 6
Authors: Peter Andrews (Barts Health NHS Trust)
Organisations: Barts Health NHS Trust
Techniques: Machine Learning
Application areas: Non-attendance Prediction, Outpatients, NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6046_Proactive_Patient_attendance_prediction/index.qmd
Links:
- Video: https://youtu.be/WNtIIWA_IYg?feature=shared
- Code: https://github.com/WX-BIU/Outpatient-DNAs
- Slides: https://drive.google.com/file/d/16aa7B_4a4i3IxZv7Cf_7GGSnVoab23i_/view?usp=drive_link

Abstract

At Barts Health NHS Trust, 12% of outpatient appointments are missed monthly, wasting over 10,000 hours of clinical resources. Missed appointments can lead to extended waiting lists and patient deterioration. This project aims to develop a machine learning model to forecast non-attendance, a patient contact capture tool, and integrate the model into enterprise reports.

Project description

Every month ~ 12% of outpatient appointments are not attended at Barts Health NHS Trust. This equates to over 10 thousand hours of clinical input, space and equipment that is not being used optimally to progress patients’ treatments or manage ongoing care.

If a patient does not attend, they could be placed back on a waiting list for an extended period which increases the risk of further deterioration or more severe progression of their condition that may result in them needing emergency care. The deterioration of the patient could have long term impacts on their future health.

This project will develop :

Machine learning model to forecast non-attendance
Patient contact capture tool
Integration of machine learning model into enterprise level reports

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: developing a machine learning model to forecast outpatient non-attendance, reducing wasted clinical capacity

“NHS 10-year plan shifts: Sickness to Prevention”: reducing the risk of patient deterioration from missed appointments.

H6-6047 — RALPulator : Predicting Robotic-assisted laparoscopic prostatectomy (RALP) operative times from patient letters

Project ID: H6-6047
Cohort: HSMA 6
Authors: Jake Wilson (Great Western Hospitals NHS Foundation Trust)
Organisations: Great Western Hospitals NHS Foundation Trust
Techniques: Machine Learning, Natural Language Processing (NLP), Streamlit
Application areas: Urology, Cancer, Surgical, NHS 10-year big bet: AI to drive productivity, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6047_RALPulator/index.qmd
Links:
- Video: https://youtu.be/ZLfEgmVCYJM?feature=shared
- Website: https://ralpulator.streamlit.app/
- Slides: https://docs.google.com/presentation/d/1rXcy9DwziUc21ukOOaWdOcQ9ZV79OGMx/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

This project is an app that reads in patient letters ahead of surgery, using Natural Language Processing techniques to extract key information from the text, and then feeds all of that into a Machine Learning model which then predicts how long the Robotic-assisted laparoscopic prostatectomy (RALP) surgery is going to take.

Project description

This project is building an app that reads in patient letters ahead of surgery, uses Natural Language Processing techniques to extract key information from the text, and then feeds all of that into a Machine Learning model which then predicts how long the Robotic-assisted laparoscopic prostatectomy (RALP) surgery is going to take.

A first version of the app has been developed and deployed, and is available here: https://ralpulator.streamlit.app/

Jake is presenting the work to the National Urological Conference in November 2024, and has been invited to present in Perth, Australia in March 2025.

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: using NLP and machine learning to extract information from patient letters and predict robotic-assisted surgery duration, supporting more efficient theatre scheduling.

“NHS 10-year big bet: robotics to support precision”: supporting the planning and resourcing of robotic-assisted laparoscopic prostatectomy surgery, a technology already transforming precision in surgical care.

H6-6048 — Identifying potential concurrent treatment areas and services that would better support patients with multiple, complex referral to treatment (RTT) pathways.

Project ID: H6-6048
Cohort: HSMA 6
Authors: Amaia Imaz Blanco (NHS England); Sean Aller (NHS England)
Organisations: NHS England
Techniques: Machine Learning, Streamlit
Application areas: Inequalities, Patient Pathways
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6048_Concurrent_treatment_areas_RTT_pathway/index.qmd
Links:
- Video: https://youtu.be/SH4geMXr8a8?feature=shared
- Slides: https://docs.google.com/presentation/d/1YXuROx59dIm9JUYWMuMbhh8A14UGov8X/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project uses machine learning to support analysis of patients who are on multiple concurrent referral to treatment (RTT) pathways, focusing particularly on healthcare inequalities. It aims to build a model that can suggest services suitable for co-location, and identify the points at which patients start having concurrent pathways.

Project description

The aim of this project is to use machine learning approaches to support analysis of patients with multiple concurrent RTT pathways, focusing particularly on healthcare inequalities.

The intention is to build a model that could identify/predict/suggest colocated services as well as points at which patients start having concurrent pathways.

H6-6052 — Geographical mapping in specialist palliative and end of life care

Project ID: H6-6052
Cohort: HSMA 6
Authors: Helen Cameron (Ashgate Hospice)
Organisations: Ashgate Hospice
Techniques: Mapping, Geographic Modelling
Application areas: End-of-life Care & Hospices, Demographics, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6052_GM_specialist_palliative_end_of_life_care/index.qmd

Abstract

This project aims to use geographic mapping with national health and census data to assess if we are caring for a fair represent the population. It will map some characteristics to include cancer/non-cancer status and protected characteristics like gender, sexual orientation, religion, and ethnicity. This understanding can support funding for specific areas and target referrals from underrepresented groups.

Project description

Using national health and census data, the aim of this project is to use geographic mapping to show if we are caring for a fair representation of our population. The two populations to be mapped would be based on the same characteristics to include cancer/non-cancer status, protected characteristics such as gender, sexual orientation, religion, ethnicity.

This would help us to understand our population and could be used to support funding for specific geographical areas or targeting referrals from underrepresented groups.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: mapping demographic and protected characteristics against hospice referral patterns to ensure fair representation and identify underserved groups who may benefit from earlier palliative referral.

H6-6055 — DES Modelling of The Hyperacute / Acute Stroke Pathway - Patient and Economic Outcomes

Project ID: H6-6055
Cohort: HSMA 6
Authors: John Williams (Maidstone and Tunbridge Wells NHS Trust)
Organisations: Maidstone and Tunbridge Wells NHS Trust
Techniques: Discrete Event Simulation (DES), Streamlit
Application areas: Stroke, NHS 10-year plan shifts: Sickness to Prevention
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6055_DES_modelling_Hyperacute_Acute_Stroke_Pathway/index.qmd
Links:
- Video: https://youtu.be/ThltRNDt9k8?feature=shared
- Article: https://arc-swp.nihr.ac.uk/news/transforming-stroke-care-through-simulation-how-one-hsma-graduates-model-could-save-over-2-million-annually/
- Code: https://github.com/jfwilliams4/des_stroke_project
- Slides: https://docs.google.com/presentation/d/18iYB7-1nJOU_3Nr0gHVSPVSsEy-VDgdz/edit?usp=drive_link&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

Stroke prevalence in the UK is forecasted to increase by 40-60% from 2021 to 2030, straining hospitals and society. This project aims to develop a discrete simulation model to optimise the Hyperacute/Acute stroke pathway, improving patient outcomes, reducing costs, and enhancing economic benefits. It will analyse variables like staffing and operating hours

Project description

In the UK the prevalence of stroke among the population is set to increase, with some forecasts putting this between 40-60% from 2021 to 2030. This increase will not only put pressure on hospitals but also on all of society.

The Hyperacute stroke setting is an extremely time critical. The quicker treatment can be delivered the better a patient’s outcome. In the Acute setting, therapy takes centre stage in rehabilitating patients, with the initial few weeks being the most impactful for the patient’s recovery.

Post stroke, patients can suffer with extreme changes to their function and wellbeing. The more severe the post stroke disability the more intensive and expensive care is needed.

The aim of this project is to develop a discrete simulation model to observe the potential effects of differing variables (such as staffing / operating hours of pathways and services) within the Hyperacute / Acute stroke pathway have on the following :

Patient flow and medical outcomes of those patients
Direct costs and savings made to the stroke unit itself
The health economic costs / benefits associated with any change

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: modelling staffing and operating hours in the time-critical hyperacute stroke pathway, where faster treatment directly improves long-term patient outcomes and reduces the need for intensive ongoing care - so not preventing the strokes, but preventing bad outcomes through adequate resourcing that allows timely access to the right people and treatments.

H6-6056 — Redrawing North West Ambulance dispatch Boundaries

Project ID: H6-6056
Cohort: HSMA 6
Authors: Charlotte Anderson (North West Ambulance Service NHS Trust); Dorota Scott (North West Ambulance Service NHS Trust); Kristian Boote (North West Ambulance Service NHS Trust)
Organisations: North West Ambulance Service NHS Trust
Techniques: Geographic Modelling
Application areas: Ambulance
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6056_Redrawing_NW_Ambulance_dispatch_boundaries/index.qmd
Links:
- Video: https://youtu.be/BONCgiiFfHc?feature=shared

Abstract

North West Ambulance Service (NWAS) has three dispatch suites in Manchester, Preston, and Liverpool, with dispatch areas mainly based on postcodes. Changes in the landscape and increased emergency demand have led to unequal resource distribution, causing delays. This project aims to generate evidence to redraw boundaries to ensure equitable resource allocation across the three areas.

Project description

Within North West Ambulance Service (NWAS), there are three dispatch suites which are based in Manchester, Preston and Liverpool. Each of the sites has a number of dispatch areas which are split geographically, based mainly on postcodes, for the purposes of dispatching emergency resources.

The geographical areas have not been reviewed for many years and in that time, there have been changes to the landscape of the areas. There have also been hospital and ambulance station closures and relocations. Over the years, there have been increases in the number of emergency ambulances to meet the demand.

Today’s issue is that the dispatch areas do not have parity in terms of how many emergency resources each are controlling and in how many allocations are made by each dispatcher. This has resulted in some dispatch areas being overwhelmed with incidents to allocate which ultimately leads to delays in responding to patients.

The aim of this project is to generate evidence to redraw these boundaries to make it more equitable across the three areas.

H6-6058 — Building a Machine Learning tool to predict Did Not Attend (DNA) events

Project ID: H6-6058
Cohort: HSMA 6
Authors: Barney Rumbold (Blackpool Teaching Hospitals NHS Foundation Trust)
Organisations: Blackpool Teaching Hospitals NHS Foundation Trust
Techniques: Machine Learning, Streamlit
Application areas: NHS 10-year big bet: AI to drive productivity
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6058_ml_tool_to_predict_dnas/index.qmd
Links:
- Code: https://github.com/BarnabyRumbold/logistic_regression_streamlit_app

Abstract

The project develops a machine learning tool that estimates the likelihood of outpatient Did Not Attend (DNA) incidences across different services and demographics. The tool could then be shared with providers to give guidance on how to reduce DNAs.

Project description

This project is developing a machine learning-based tool that looks at the likelihood outpatient Did Not Attend (DNA) incidences across different services/demographics.

The tool could hopefully be shared with providers to provide guidance on how to reduce DNAs.

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: building a machine learning tool to predict outpatient non-attendance likelihood across services and demographics, supporting providers to reduce wasted clinical capacity.

H6-6060 — Optimising Same Day Emergency Care (SDEC) Resourcing

Project ID: H6-6060
Cohort: HSMA 6
Authors: Hannah Thould (University Hospitals Bristol and Weston NHS Foundation Trust)
Organisations: University Hospitals Bristol and Weston NHS Foundation Trust
Techniques: Discrete Event Simulation (DES), Emergency Departments
Application areas: NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6060_optimising_resourcing/index.qmd
Links:
- Video: https://youtu.be/Yck3vOioRdQ?feature=shared
- Code: https://github.com/hthould/acute_take_des
- Slides: https://docs.google.com/presentation/d/1fGS1mKo8BrF_I4wy5AAzjDsdETnUaFnE/edit?usp=sharing&ouid=104927246423235110137&rtpof=true&sd=true

Abstract

The project uses Discrete Event Simulation to model Same Day Emergency Care (SDEC) and Emergency Department pathways at University Hospitals Bristol and Weston NHS Foundation Trust. It aims to establish the optimum parameters for those pathways, the effect of different doctor numbers, and the optimum size for SDEC.

Project description

This project will model Same Day Emergency Care (SDEC) and Emergency Department (ED) pathways at University Hospitals Bristol and Weston NHS Foundation Trust.

The project aims to use Discrete Event Simulation to ascertain

optimum parameters for SDEC and ED pathways
the effect of different doctor numbers
the optimum size for SDEC

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: modelling optimum SDEC sizing and staffing to support same-day emergency care, helping patients avoid unnecessary overnight hospital admission.

H6-6061 — Analysis and forecasting of referrals into hospital

Project ID: H6-6061
Cohort: HSMA 6
Authors: Andrew Sharrock (The Walton Centre NHS Foundation Trust)
Organisations: The Walton Centre NHS Foundation Trust
Techniques: Forecasting
Application areas: Neurology, Demand & Capacity, Referrals
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6061_analysis_forecasting_hospital_referrals/index.qmd

Abstract

The project uses Python and forecasting approaches to analyse the growth in referrals into the hospital.

Project description

This project will use Python and forecasting approaches to analyse growth in referrals into the hospital.

H6-6062 — Developing a primary care load management tool

Project ID: H6-6062
Cohort: HSMA 6
Authors: Chris Lewis (The Park Medical Practice)
Organisations: The Park Medical Practice
Techniques: Discrete Event Simulation (DES)
Application areas: Primary Care (GP), NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6062_primary_care_load_management_tool/index.qmd
Links:
- Video: https://youtu.be/AUNmuPFnq4c?feature=shared
- Code: https://github.com/clexp/Primary_Care_Load_Management_Tool

Abstract

A GP practice had peak call wait times of 60 minutes. Adding more staff improved this, but lower utilisation was then seen during quieter periods. This project will model the telephone system and call handlers’ workload to optimise resourcing, producing a web app to test different scenarios.

Project description

A GP practice serving 35,000 patients experiences peak call wait times of 60 minutes. While increasing staffing levels has reduced these wait times, this approach can result in periods of lower utilization during quieter times.

This project aims to develop a comprehensive model of the telephone system and call handler workload patterns. The goal is to create an analytical tool that can evaluate telephone resource allocation and identify opportunities for operational optimization.

A web-based application will be developed to enable testing of various staffing scenarios and resource allocation strategies.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: modelling GP telephone system and call handler capacity to reduce peak call wait times, supporting timely access to primary care and ending the “8am scramble” for appointments.

H6-6063 — Modelling the Talking Therapies clinical pathway using a Discrete Event Simulation

Project ID: H6-6063
Cohort: HSMA 6
Authors: Heath McDonald (Lancashire & South Cumbria NHS Foundation Trust)
Organisations: Lancashire & South Cumbria NHS Foundation Trust
Techniques: Discrete Event Simulation (DES)
Application areas: NHS Talking Therapies (Formerly IAPT), NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_6/H6_6063_modelling_talking_therapies_clinical_pathway_using_des/index.qmd
Links:
- Code: https://github.com/MightyAtom220474/IAPT-Pathway-DES
- Paper: https://drive.google.com/file/d/15AsF2U4BnsXbbGQRxIcFkDW130j4JiaF/view?usp=drive_link

Abstract

The NHS Talking Therapies programme, supports NICE guidelines for treating anxiety and depression. The aim is to develop a Discrete Event Simulation to model patient flow through the new clinical pathway. This project will identify potential waiting list build-ups due to increased referrals into Talking Therapies.

Project description

According to the Office of National Statistics (ONS), almost one fifth of adults in the UK experience anxiety or depression. The Depression Report (Layard, 2005) identified psychological therapies as vital in the treatment of anxiety and depression. In October 2007, the UK government announced a large-scale programme to improve psychological health in the population: Improving Access to Psychological Therapies (IAPT).

Following 15 years of delivery, the programme underwent a national rebrand and was re-named in early 2023 as “NHS Talking Therapies for anxiety and depression”. The NHS Talking Therapies programme supports the NHS in implementing National Institute for Health and Clinical Excellence (NICE) guidelines for the treatment of those experiencing common mental health problems such as depression and a range of anxiety disorders, providing Talking Therapies services within the framework of a stepped care approach.

Lancashire & South Cumbria NHS Foundation Trust recently reviewed its pathway in line with National Recommendations for how Talking Therapies services should be delivered.

The aim of this project is therefore to develop a Discrete Event Simulation to model the flow of patients through this new Talking Therapies clinical pathway incorporating the entire clinical pathway – Screening & Assessment, Step 2 and Step3, and modelling patient flow both within and between all the elements of the pathway in an effort to identify any areas of potential waiting list build up resulting from the increased numbers of referrals into Taking Therapies that are being received.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: modelling patient flow through the NHS Talking Therapies pathway to identify potential waiting list build-up, supporting timely access to community-based mental health treatment.

H6-6065 — Automating injury coding using language models

Project ID: H6-6065
Cohort: HSMA 6
Authors: James Lai (Imperial College Healthcare NHS Trust)
Organisations: Imperial College Healthcare NHS Trust
Techniques: Natural Language Processing (NLP), Machine Learning
Application areas: Clinical Coding, NHS 10-year big bet: AI to drive productivity, NHS 10-year big bet: Data to Deliver Impact
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6065_automating_injury_coding_using_language_models/index.qmd

Abstract

Traumatic injuries are common in emergency care and a leading cause of death and disability in working-age individuals. Patients undergo clinical assessments, blood tests, and CT imaging upon arrival. Major trauma centres (MTCs) are funded based on injury severity, requiring accurate coding. The HSMA project aims to train a language model to generate injury codes from free-text radiology reports.

Project description

Traumatic injuries are a very common presentation to emergency care and the leading cause of death and disability in the working-age population. Patients can sustain multiple injuries to varying locations and of varying severity.

On arrival at the emergency department. The injured patient will have a clinical assessment and blood tests to evaluate the degree of injuries. Once stabilised after arrival in the emergency department, the acutely injured patient will require CT imaging to guide clinical management. Clinical imaging is where most of the injuries are detected.

Major trauma centres (MTCs) are funded to deliver activity. The injury casemix and severity for each MTC is reported centrally, and centres are funded for the delivery of major trauma activity, with an increased tariff for patients with a higher injury severity score (ISS), reflecting the severity of injury. Therefore, accurate coding is required to reflect the injury case mix a centre receives and attract the appropriate tariff for the delivery of major trauma activity.

The aim of the HSMA project is to train a language model to take free-text radiology report inputs and generate an injury code based on the free-text information provided.

10-year plan Alignment

“NHS 10-year big bet: AI to drive productivity”: training a language model to generate injury codes from free-text radiology reports, automating a task that currently requires manual clinical coding effort.

“NHS 10-year big bet: Data to Deliver Impact”: creating additional structured data from unstructured data, supporting further work down the lines

H6-6066 — Evaluating the Impact of Community Diagnostic Centres on Health Inequalities and Patient Access to Diagnostic Services

Project ID: H6-6066
Cohort: HSMA 6
Authors: Felix Mukoro (NHS England)
Organisations: NHS England
Techniques: Geographic Modelling, Streamlit, Python
Application areas: Community Diagnostic Centres (CDCs), NHS 10-year plan shifts: Sickness to Prevention, NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6066_evaluating_impact_of_community_diagnostic_centres/index.qmd

Abstract

Community Diagnostic Centres (CDCs) aim to expand diagnostic capacity and improve access, but their equitable distribution is unclear. This project will assess CDCs’ impact on patient access and health inequalities by analysing socio-demographic, geographic, and utilisation data.

Project description

The introduction of Community Diagnostic Centres (CDCs) aims to expand capacity and bring diagnostics closer to communities, ideally reducing waiting times and improving access. However, whether these benefits are being equitably distributed remains unclear.

Evidence is needed to quantify the true effect of these centres, including their impact on groups that might experience challenges in accessing timely diagnostic services (e.g., those in deprived areas, rural communities, or with limited transport).

This project seeks to investigate whether CDCs have effectively improved patient access, and reduced health inequalities. By combining socio-demographic, geographic, and utilisation data, this project aims to identify areas and populations still underserved by diagnostics and propose strategies for equitable service provision.

This project and output will feed into the overall CDC Programme Evaluation/Review which was recently commissioned by the CDC Programme national team and the Department of Health and Social Care (DHSC).

The aims of this project are to : - Quantify how the phased opening of CDCs has changed overall diagnostic capacity and waiting times. - Assess how the roll-out of CDCs has influenced access to diagnostic services across diverse socio-demographic groups and geographical areas. - Determine whether health inequalities in diagnostic utilisation or waiting times have lessened post-CDC introduction.

10-year plan Alignment

“NHS 10-year plan shifts: Sickness to Prevention”: assessing whether Community Diagnostic Centres have equitably improved access and reduced health inequalities, supporting the plan’s aim to narrow the gap in healthy life expectancy.

“NHS 10-year plan shifts: Hospital to Community”: evaluating how Community Diagnostic Centres bring diagnostic capacity closer to local populations, reducing waiting times outside of acute hospital settings.

H6-6067 — Applying and manipulating identification rules for specialised services

Project ID: H6-6067
Cohort: HSMA 6
Authors: Emma Keeley (Arden and GEM CSU)
Organisations: Arden and GEM CSU
Techniques: Streamlit, Python
Application areas: Specialised Services, SUS Dataset
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6067_applying_manipulating_id_rules_specialised_services/index.qmd

Abstract

Specialised service lines are identified by the Prescribed Specialised Services tool, a black box whose hierarchical rules change each year and cannot be adjusted. The project builds a tool that applies and manipulates those rules against SUS data, showing movement between service lines and the resulting changes in patient demographics and travel times.

Project description

Specialised service lines are identified using Prescribed Specialised Services (PSS) Tool and returned as a field within the SUS dataset. Each year there are changes to the rules and a new tool is provided. The PSS tool is a black box, so there is no access to the manipulate the rules within the tool. The IR rules are hierarchical, a patient can meet the criteria for multiple service lines but is assigned the service line that is highest in the hierarchy. Due to the hierarchy, a change in one rule can cause movement between specialised / non-specialised but also changes within final service line. SUS inpatient and outpatient tables combined contain in excess of 2 billion rows, understanding changes across service lines requires processing all of the data and is therefore slow analysis in SQL.

The aim of this project is to build a tool that will allow application and manipulation of the IR rules to the SUS datasets. The output will show movement between specialised / non-specialised, movement between service lines and changes in delegated/non-delegated services. As SUS is a patient level dataset there is opportunity to look at changes in patient demographics caused by amendments to the IR rules. As many of the services lines have associated provider eligibility lists, the output should also show changes in travel time for patients. There is potential to be able to analyse the elements of the rule criteria that have been met to assign the service code, this will help determine if there are redundant elements of the rules.

H6-6069 — Generating wordclouds from referral information

Project ID: H6-6069
Cohort: HSMA 6
Authors: Rebecca Marshall (University Hospitals Coventry and Warwickshire Trust)
Organisations: University Hospitals Coventry and Warwickshire Trust
Techniques: Natural Language Processing (NLP)
Application areas: Referrals
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6069_wordclouds_referral_information/index.qmd

Abstract

Referral order forms often contain large paragraphs of free text in the referral reason field. The project uses Natural Language Processing techniques to extract the information most commonly recorded there, so that the service can understand the most common reasons for referral.

Project description

The aim of this project is to use Natural Language Processing techniques to extract commonly used information in the referral reason of referral order forms.

Often large paragraphs of free text are written here, but the service would find it useful to understand what are some of the most common reasons for referral.

H6-6070 — Using Machine Learning techniques to predict Hospital Readmissions

Project ID: H6-6070
Cohort: HSMA 6
Authors: Rebecca Marshall (University Hospitals Coventry and Warwickshire Trust)
Organisations: University Hospitals Coventry and Warwickshire Trust
Techniques: Machine Learning, Explainable AI
Application areas: NHS 10-year plan shifts: Hospital to Community
Status: Active
Public code repository: No
Source file: previous_projects/hsma_6/H6_6070_machine_learning_hospital_readmissions/index.qmd

Abstract

The Trust averages 480 emergency 30-day readmissions monthly. This exploratory project aims to use Machine Learning to predict readmissions and identify key factors. It will assess the accuracy of different approaches using current Trust data, potentially guiding future work in this area.

Project description

Each month, the Trust has, on average, 480 Emergency 30-day readmissions (7.7%). My aim is to use Machine Learning techniques to try predict who will get readmitted and identify any key factors which make it more likely for someone to be readmitted.

At this stage, it is an exploratory project that will look more at what level of accuracy is possible to achieve with different approaches from the data currently held by the trust, and may inform future direction of work in this area.

10-year plan Alignment

“NHS 10-year plan shifts: Hospital to Community”: using machine learning to predict which patients are likely to be readmitted within 30 days and identify key contributing factors, supporting more targeted preventative follow-up and preventing patients from needing to return to hospital.

H1-1100 — Geographic modelling of resource utlitisation at Devon Air Ambulance

Project ID: H1-1100
Cohort: HSMA Alumni
Authors: Hannah Trebilcock (Devon Air Ambulance); Richard Pilbery (Yorkshire Ambulance Service Trust)
Organisations: Devon Air Ambulance
Techniques: Geographic Modelling, Forecasting, Discrete Event Simulation (DES)
Application areas: Ambulance, Air Ambulance, Patient Transport, Waiting Times
Status: Active
Public code repository: No
Source file: previous_projects/hsma_alumni/H1_1100_Geographic_modelling_of_resource_utilisation_at_DAA/index.qmd
Links:
- Code: https://github.com/RichardPilbery/DAA_DES

Abstract

The project uses geographic modelling to examine the utilisation of resources at Devon Air Ambulance. It models current demand and runs what-if scenarios in order to optimise service delivery and provide an evidence base for operational decision making.

Project description

A geographical modelling project on the utilisation of resources at Devon Air Ambulance. Modelling current demand and conducting what if scenarios to optimise service delivery & provide an evidence base for operational decision making.

H5-5100 — eFIT: Extra funding allocation - inequality tool

Project ID: H5-5100
Cohort: HSMA Alumni
Authors: Peter Saiu (NHS England)
Organisations: NHS England
Techniques: Streamlit
Application areas: Inequalities, Costs
Status: Active
Public code repository: Yes
Source file: previous_projects/hsma_alumni/H5_5100_Extra_funding_allocation_inequality_tool/index.qmd
Links:
- Video: https://youtu.be/YbmiEilW9I8
- Code: https://github.com/pete4nhs/eFIT-tool
- Website: https://www.heec.co.uk/resources/extra-funding-allocation-inequality-tool-efit/
- Slides: https://youtu.be/YbmiEilW9I8?feature=shared

Abstract

The project addresses the lack of national guidance for allocating extra primary care funding by ICBs. It proposes using an equation based on deprivation scores and local needs. A Streamlit web-app tool will help ICBs allocate funds more equitably, considering various indicators and demographics, ensuring a fair distribution and reducing inequalities.

Project description

Whilst core funding for primary care is well regulated (by using a variant of the Carr-Hill formula), there is no national guidance/methodology to allocate extra funding for primary care by Integrated Care Boards (ICBs) Every ICB has many programme directors/project managers for the various disease areas who can split the money at their own discretion who might not take into consideration available data and might just split the money equally (by population size) - effectively widen inequalities even further.

The project proposes to use an equation from our Inequality strategy review which splits the money based on deprivation score and local needs. For example, if a cancer team receives extra funding for FIT screening, I calculate the allocation of extra funding to GP practices based on bowel cancer screening uptake data and give more to those who have low uptake (in a proportional way).

Many indicators can be used, so a team focusing on high-intensity users will use data on high-intensity users etc. However, even if programme managers were aware of the equation, they might not know how to calculate the allocation and might need analysts to do it for them.

I’m creating a tool using the user-friendly Streamlit web-app where a user would input the amount of funding, select the ICB, specify the demographics (age and gender for intervention) and tweak the indicator and the weights (e.g. to give more based inequality or prevalence of disease etc).

Upon hitting the ‘Calculate’ button the result is instantaneous and you can download the table with the calculated allocations.By making this tool available to all ICBs nationally, we can establish a general consensus on how ICBs can promptly split extra funding to primary care taking into consideration inequalities and local needs.