HES data through Signals From Noise (sfn) Business Intelligence Platform

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Datasets: − HES-ID to MPS-ID HES Accident and Emergency; − Hospital Episode Statistics Accident and Emergency (HES A and E)

Objective for processing

[3 paragraphs unchanged] Article 6(1)(f) - processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party. For this application, Lightfoot have determined that complying with Article 6(1)(f) legitimate interest for the processing of data requested, and subsequent processing is necessary to ensure service planning and outcomes are optimised for the purposes health services by using statistics to analyse service provision and monitor for unintended consequences of Legitimate interests as pursued by a controller, except where such interests are overridden by the interests or fundamental rights and freedoms redesign that could affect quality of the data subject which require protection of personal data, in particular, where the data subject is a child. care. Lightfoot have determined that complying with Article 6(1)(f) legitimate interest for the processing of data is necessary to ensure service planning and outcomes are optimised for health services by using statistics to analyse service provision and monitor for unintended consequences of redesign that could affect quality of care. [1 paragraph unchanged] Article 9(2)(j) - processing is necessary for archiving purposes in the public [45 words unchanged] to safeguard the fundamental rights and the interests of the data subject. The processing is necessary for the performance of a task carried out in the public interest for the cases of planning and performance management to improve patient outcomes. The processing is necessary for the performance of a task carried out in the public interest for the cases of planning and performance management to improve patient outcomes. [36 paragraphs unchanged] Clients are only approved if they are an NHS organisation or academic [74 words unchanged] be noted on Lightfoot’s website, and Lightfoot will retain minutes and NHS Digital England will audit compliance with minutes/meeting procedure and the terms of reference as part of this agreement. Lightfoot Solutions are the Data Controller who also processes the data. Lightfootare Lightfoot are permitted to process the data to provide the outputs detailed below. [5 paragraphs unchanged] Lightfoot are currently working with NHS England and Mid Yorks NHS Hospitals Trust. [1 paragraph unchanged]

Processing activities

[7 paragraphs unchanged] For some customers within the customer groups stated above, Lightfoot will provide [79 words unchanged] full training for all such users. All users must comply with NHS Digital's England's Hospital Episode Statistics (HES) Analysis Guide. [22 paragraphs unchanged] Data provided by NHS Digital England under this agreement will not be linked with any other record level data.

Expected output

[15 paragraphs unchanged] The product includes a user configurable demand model to help systems predict acute urgent care activity and occupancy levels to support winter planning for 2022/23. 2023/24. [3 paragraphs unchanged] • Understand our baseline patient needs and problem we are trying problems needed to fix be fixed • Understand the improvements we were making that have been made and ensure our the outcomes were delivering value [18 paragraphs unchanged]

Expected measurable benefits

Public health decision making continues to become increasingly complex, and the use of data has become essential to this process. To provide the ability for Healthcare Organisations to view their data through a different lens, the implementation of Integrated Care Systems has enhanced the need for a consistent approach to support integration, evaluation, planning and resource investment. By supporting this approach, we can identify opportunities that will improve the efficiency and effectiveness of a health system at a local level to monitor population health and to target interventions. At a national level, this data can be used for effective resource allocation and short-, medium- and long-term planning. These wider public health benefits can include reduced wait times for treatment, reduced admissions to hospital, reduced length of stay upon admission to hospital and improved aftercare. This approach will also benefit the wider public as viewing data in this way will enable faster responses to new threats to public health, such as a reoccurrence of the COVID-19 pandemic. The benefits highlighted below are measured in outcomes tracked through the sfn Viewer. The workstreams are ongoing and use the monthly updates of HES data to monitor improvement and explore changes in variation. Target dates are ongoing as the work extends through the approaching Winter. The high-level benefits for the public, and their health listed both above, and those detailed below are measured through outcomes tracked through the sfn Viewer. Benefits Benefit to our clients is client was the ability to identify, from processing the HES data using sfn, opportunities to reduce bed days as part of winter preparedness and ongoing system planning with a focus on maintaining Elective recovery since the COVID-19 pandemic. The workstreams are ongoing and use the monthly updates of HES data to monitor improvement and explore changes in variation. Target dates are ongoing as the work extends through the approaching Winter. Our work is focused on identifying key constraints, variation, and bottlenecks to improve the flow of a healthcare system. [1 paragraph unchanged] sfn provides an opportunity to view planned care alongside urgent and emergency care (UEC) in a single view, overlaid with our Lightfoots Population Cohort analytics to segment patients and identify which cohorts are having the biggest impact on the system, thus supporting the healthcare provisions on the public. system. [12 paragraphs unchanged] The sfn analysis has identified cohorts of patients at high risk of an extended admission based on their clinical [27 words unchanged] of Winter planning could significantly reduce pressure on the urgent care system. [3 paragraphs unchanged] Lightfoot’s Lightfoots solutions focus on the changing rate, rather than the absolute value to [5 words unchanged] to suffer operational challenges if the projected demand growth is not mitigated. [3 paragraphs unchanged] The ‘PBC code’ is needed to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. [2 paragraphs unchanged] Other Benefits The ‘PBC code’ is needed to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. Lightfoot have been commissioned to support several ambulance trusts in analysis of patient journey data with ambulance trusts to review clinical models in order to deliver the local urgent care strategy and develop a model to treat more people at home and refer to local community services where appropriate. Lightfoot is in discussion with these ambulance services to support redesign of their clinical model to improve patient outcomes. Provider A&E data is now submitted via the ECDS sfn measures are mapped across the ECDS standard and the legacy A&E CDS which enables long term trend and variation analysis covering the migration period. This is particularly helpful for strategic population-based analysis where long sample periods improve the accuracy of projections. The national level data available in HES provides sufficiently large sample sizes for meaningful evaluation of services designed to impact small numbers of patients with relatively rare characteristics or low volume services. The HES data provides a large control group for this type of evaluation which often increases the confidence level of any significant results.

Benefits reported

[1 paragraph unchanged] 1 Lightfoot worked with an acute provider and their commissioner clinicians to look [10 words unchanged] used to understand the impact of new frailty clinics in the community. The Trust have used the HES data to consider the definition of outcome measures for their acute frailty pathways and want to [21 words unchanged] of a continuous improvement framework that will require a daily data feed. NHS England and ICSs in South East have experienced the following benefits listed below from the sfn analytics platform which has measures and configurations available out of the box plus localised configurations that have been deployed as part of the earlier phases of the work. 2 Demand growth benchmarking and Recovery The product enables systems to compare the relative urgent care demand growth within their population to identify areas of high growth. Users can drill down into populations (to GP practice level) and understand whether changes in the demographic profile of the population correlate with demand pressure on the urgent care system. The interface is designed to help systems identify areas where demographic change does not explain high levels of demand growth. The project highlighted specific populations where demand growth was high and proactive intervention could potentially avoid hospitalisation, examples were highlighted in: • Frimley ICS • Sussex Health and Care Partnership 3 Supporting Winter Planning The product includes a user configurable demand model to help systems predict acute urgent care activity and occupancy levels to support winter planning for 2022/23. This model accounts for normal seasonal demand pressures and enables users to configure scenarios based on a ‘return to new normal’ following the impact of COVID. Benefits could be shown using demographics and non-demographics to highlight pre COVID demand and growth summary. How data enabled a GP Practice in Kent to develop, measure and accelerate the delivery of improvement using sfn Lightfoot. sfn helped: • Understand baseline patient needs and problems Lightfoot are trying to fix • Understand the improvements and ensure Lightfoot's outcomes were delivering value • Benchmark against neighbouring PCN and the wider Kent system. • Demonstrate the evidence of improvement ‘Making Data Count’ to colleagues utilising quantitative measures • This GP practice continues to show improvement and uses the HES data to quantify this to support other PCNs to adopt the same approach Benefits Some early feedback from clinicians, both in primary and secondary care: • Patients, families and carers spend more time in their place of choice – preferably home • Health and Social Care workforce has more time to address the right individuals • Increased ambulance capacity not conveying those who could be cared for at home (or in nursing home) • Better system flow creates capacity to address the medium to long term Elective Care challenge • Significant financial benefits (avoidance/reduction) through beds being occupied by those patients who most require acute care • Creates transformation fund to further improve and expand primary and community services to replicate success across the wider East Kent health and social care system • Improve system safety through professional teams having quality time to proactively measure and monitor patients • Innovative and high performing systems attract the additional workforce required to address the current / future challenge • Improves morale across the health and social care system • Accelerates collaborative working 4 NHS England and ICSs in South East have experienced the following benefits listed below from the sfn analytics platform which has measures and configurations available out of the box plus localised configurations that have been deployed as part of the earlier phases of the work. [1 paragraph unchanged] • Central to moving towards a patient centric model of healthcare, based around integrated healthcare systems, the ambition is to: [1 paragraph unchanged] • To increase the ‘transparency’ of organisational boundaries which may be inhibiting collaboration and consequently patient centered centred improvement • A key enabler to this strategic change is aligned data and a strong informatics capability, which is centered centred on people not organisations - for whole populations; cohorts of that population; [14 words unchanged] readily into the hands of key decision makers, planners and most importantly our clinicians • Analysing and understanding current and past data has value, but being able to forecast forwards in a statistically robust way will support us Lightfoot to make optimal use of the constrained resources across and within systems, optimising outcomes for our populations • For systems this would represent a notable shift in how we Lightfoot seek to understand the operation of our healthcare services and the assurance of quality care that follows [2 paragraphs unchanged] In September 2021 Lightfoot was engaged by the Southeast Region of NHS England to use the HES dataset to provide insights into the effect of the Covid pandemic on the delivery of both Urgent and Emergency Care and Planned Care and to identify the factors affecting both the demand for hospital-based care and the pressure on bed capacity and to identify the interventions that would have the most impact in helping to mitigate UEC demand, 5. Providing a flow based approach to understanding UEC & elective demand and show population-based variation at ICS, Place and PCN level The insights from the analysis were shared with the 6 ICSs and the 18 Acute Trusts across the South-East Region to support their winter readiness planning. This analysis showed that the growth in the demand for urgent care in the post pandemic period was in most cases in line with or below the pre-pandemic trend level, but that there were particular groups of patients who were placing unusual pressures on different systems. The key finding of this analysis was the identification of a cohort of ‘High Risk’ patients who are aged over 50 yrs and have previously had a 14-day length of stay. These patients account for around 2 per cent of the registered GP population in each of the ICSs in the South-East, but account for more than 50% of the occupied beds in the region. This insight has enabled local systems to focus their out of hospital admission avoidance planning and resources on this group of patients with clear benefits in terms of a reduction in the rate at which these patients are being readmitted and a corresponding reduction in the beds occupied by this cohort of patients. Examples where this approach has been adopted to good effect include an admission avoidance program undertaken in Banbury by the Oxford region of the BOB ICB and in East Kent in the Kent and Medway ICB. In addition, Lightfoot has also used the HES data to support the Clinical Leader of the Year for 2021, to evidence the effectiveness of the model of care for older people that he has developed at his practice in Thanet. The success of Lightfoot’s approach in delivering benefits from the use of the HES data has been recognised by the leadership team in the South-East Region who have recently extended Lightfoot’s contract to support the ICBs in the region in recovering from the effects of the pandemic . The South-East region has been working towards improving their access to BI resources with our help. By leveraging the robustness of Statistical Process Control (SPC), we have been able to provide the South-East region with powerful insights into the behaviour of real-world processes and patient pathways. An always-on visual control & monitoring system, it has provided robust evidence to support decisions about change or improvement, identifying hidden insights from our processes. They have been able to measure and monitor the performance of any combination of processes from multiple organisations along the full-service pathway. Our dashboards have delivered up-to-the-minute information and reports to people across the organisation. In this way, process issues can be quickly identified, and targeted actions can be decided upon to improve outcomes. With sfn, the visual modelling of key aspects has been achieved in a variety of ways, such as : Annual, monthly, weekly and daily cyclicity of data / Trends and projections / Anomalies and unusual behaviour. This is allowing UK health organisations to compare their performance and outcomes with other similar UK health organisations to improve patient outcomes. Using the data provided, NHS England have been provided with the tools to review their modelling structures by being able to consistently review the same data across the South-East Region. By looking deeply into the data, in the viewers we have created they have been able to understand what drivers are causing increased wait times for ambulances. By reviewing the data to understand the differences it has highlighted Portsmouth as delivering differently to other areas of the South-East. A hospital base process is now being initiated to focus on the problem. This will have a wider impact for the communities involved and supports the focus on improving patient outcomes for the benefit of the public. Benefits from Winter Planning initiatives sfn Population Cohort analysis to identify opportunities and optimal solutions to mitigate acute capacity pressures • Identified patients at high risk of an extended admission based on their previous in-hospital behaviour and LOS o This is a small group per GP practice but there is a high utilisation rate so targeting this group as part of winter planning could significantly reduce pressure on the urgent care system • Empirical risk stratification to support high impact interventions, such as the use of virtual wards or high-risk adult cohort interventions • Advanced cohort analysis provided a unique insight into how different patient groups move through the system, where potential demand pressures may arise and where interventions will be most pivotal in preventing queues and maintaining flow • sfn automatically predicted levels using the demand history relative to the population and gave the benefit of being able to also assess unwarranted variation within local systems • Provide analysis and feedback for current models/improvements for community step up and front door services for high-risk adults • sfn was used to test and develop intervention logic and allows users to evaluate, do future modelling and track the success of interventions • sfn is supporting development of the Integrated Care strategy by identifying the cohorts of older, at-risk patients who will benefit most from a more focused approach to personalised care planning and community-based care delivery working. • Identification of the High-Risk cohort - a group of patients, who use 30% to 60% of the beds Providing a flow based approach to understanding UEC & elective demand and show population-based variation at ICS, Place and PCN level [7 paragraphs unchanged] 6. Working with a Health Economy in Wakefield • Data-led decision-making to support winter planning • Baseline workshop- A series of interactive workshops using the HES data took place culminating in a baseline playback workshop. This ‘deep dive’ baseline session allowed senior leaders and clinicians to identify hot spots and identify key opportunities to make an impact and develop a shared vision for priority areas. • The project team used Hospital Episode Statistics (HES) data and sfn's real-time data analytics and predictive modelling capabilities to identify areas of high demand and potential service pressure during the winter months • Using the national dataset allowed a ‘quick start’ to proactive planning and resource allocation to take the system through winter whilst local inpatient (IP) and emergency department (ED) data was inputted and configured into the sfn platform. • This advanced cohort analysis provided early focus on the High-Risk Adult Cohort; patients at high risk of an extended admission based on their clinical coding for frailty and previous length of stay (LoS). Benefits • This analysis underpinned a redesign of health and social care responses to people over 50 who often have extended hospital care. • This small group of people, who have only been identified by taking a population cohort approach to hospital activity data, account for roughly 60% of unplanned inpatient bed occupancy, are also high community service users, and are likely to require long-term residential care. • Using the sfn data to identify this population, understand their system demand, forecast their future activity and design new integrated and targeted responses based on the best evidence has delivered a substantial change. • In a three month pilot, the combined approach focused on reducing admissions, accelerating discharge and intensifying community-based rehabilitation has delivered a reduction in hospital bed days that annualises to 35 beds per annum. There are also early signs of a reduction in long-term residential care requirements. These results act as an example of the benefits of integrating health and social care responses for targeted populations.

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

The data requested and subsequent processing is necessary for the purposes of the legitimate interests pursued by a controller, except where such interests are overridden by the interests or fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child. Lightfoot Solutions have a legitimate interest to ensure service planning and outcomes are optimized for health services by using statistics to analyze service provision and monitor for unintended consequences of redesign that could affect quality of care. Processing is also necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject. Lightfoot Solutions UK Ltd (Lightfoot) are an organisation who work to help healthcare organisations transition from a traditional silo-based structure to a flow-based system-wide management approach. By incorporating data from different healthcare providers, Lightfoot can measure patient outcomes across the whole pathway, linking all the services in each patient’s journey. Lightfoot are an organisation who work to help healthcare organisations transition from a traditional silo based structure to a flow-based system-wide management approach. By incorporating data from different healthcare providers, Lightfoot is able to measure patient outcomes across the whole pathway, linking all of the services in each patient’s journey. [1 paragraph unchanged] Work streams are provided with a Statistical Process Control (SPC) view of current pathways for patients and to identify key constraints, variation and bottlenecks in the various patient pathways which are used to monitor and evaluate the impact of the improvement actions and advise when necessary intervention should take place. The lawful basis for processing personal data under the UK GDPR is: Article 6(1)(f) - processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party. For this application, the data requested, and subsequent processing is necessary for the purposes of Legitimate interests as pursued by a controller, except where such interests are overridden by the interests or fundamental rights and freedoms of the data subject which require protection of personal data, in particular, where the data subject is a child. Lightfoot have determined that complying with Article 6(1)(f) legitimate interest for the processing of data is necessary to ensure service planning and outcomes are optimised for health services by using statistics to analyse service provision and monitor for unintended consequences of redesign that could affect quality of care. The lawful basis for processing special category data under the UK GDPR is: Article 9(2)(j) - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject. The processing is necessary for the performance of a task carried out in the public interest for the cases of planning and performance management to improve patient outcomes. There are no moral or ethical issues for the dissemination of this data as Lightfoot’s clients are given direct access to perform analysis on the Hospital Episode Data using the Business Intelligence Platform, that is written and hosted by Lightfoot. The client can access a suite of dimensions and measures to perform their analysis. The results of this analysis are always aggregated and automatically has small numbers supressed before being returned by the application in graph and chart formats only. Lightfoot classify the data as an aggregated statistic as the final use of the data and the audience viewing it do not have access to record level data, and therefore personal data cannot be re-identified. All processing of record level data takes place on hardware and software wholly owned and controlled by Lightfoot. Lightfoot confirm that no record level data is provided to any third-party customers. Lightfoot confirm as a result, there is very little risk of potential harm to the public for this processing. Work streams are provided with a Statistical Process Control (SPC) view of current pathways for patients and to identify key constraints, variation and bottlenecks in the various patient pathways which are used to monitor and evaluate the impact of the improvement actions and advise when necessary, if interventions should take place. [1 paragraph unchanged] Lightfoot provide the signalsfromnoise (sfn) a predictive analytics statistical tool, Business Intelligence platform, which is used by or for (where Lightfoot are providing the service) the following non-commercial organisations: NHS Providers, Commissioners, STPs, Integrated Care Systems (ICSs) (down to Place and PCN level), NHS EI, England, NHS Professional Associations and Academic Health Science Networks. [4 paragraphs unchanged] 3. The client determines the nature of any analysis and the final use and audience of that analysis – excepting that the result will be an aggregated statistic statistic. 4. Lightfoot provide the software, platform, storage storage, and access to the data. In all HES customer use cases using statistical analysis analysis, the drill down platform will automatically suppress small numbers before being presented [18 words unchanged] Lightfoot confirm that no record level data will be provided to any third party customers or Lightfoot internal consultants and analysts. third-party customers. [1 paragraph unchanged] 1. Providing access to summary and statistical analysis of patient data to [7 words unchanged] greater understanding of patient activity and flow to support the following activities in order to improve health provision: provision using the Statistical Process Control: [4 paragraphs unchanged] 2) E. Providing access to summary and statistical analysis of patient data to NHS commissioning organisations England and ICSs to support healthcare planning and service redesign, using the Statistical Process Control (SPC) view to: a. a) Provide a view of current patient pathways and to identify key constraints, variation variation, and bottlenecks in the various patient pathways; pathways b. b) Monitor and evaluate the impact of the improvement actions. actions 3) c) Providing access to summary and statistical analysis of patient data to Ambulance Trusts NHS organisations to support service improvement programmes. programmes 4) Providing access to summary and statistical analysis of patient data to the Association of Ambulance Chief Executives (AACE) to enable and support national improvement programmes by using HES data to demonstrate outcomes of patient cohorts taken to hospital via the Urgent & Emergency Care pathway and conveyed by ambulance. d) Identifying opportunities to reduce bed days as part of winter preparedness and ongoing system planning with a focus on maintaining Elective recovery e) Planning and modelling capability to support systems to address the potential pressures this winter whilst maintaining Elective Recovery f) sfn Population Cohort analysis to identify opportunities and optimal solutions to mitigate acute capacity pressures g) Providing a flow-based approach to understanding urgent and emergency care (UEC) & elective demand and show population-based variation at ICS, Place and PCN level h) Enabling ICSs and Trusts to evaluate the effect of the COVID-19 pandemic on the demand for both Urgent and Emergency and Planned care and to compare this with the pre-pandemic trend levels of activity i) Providing ‘what if’ models that evaluate the impact over time of interventions such as Virtual Wards to reduce UEC demand j) Providing a pathway methodology to enable ICSs and Trusts to evaluate the progression of Planned Care demand and waitlists over time and to assess the progress that is being made in addressing the post-pandemic backlog in Elective activity. Justification for the requested data sets : 1. To allow a view of patient outcomes for those attending the Emergency Department by Ambulance both the Emergency Department and Admitted Patient Care data is required. This allows users to view the whole patient journey to show outcomes of those patients who arrive via an ambulance. 2. Lightfoot's work with Health economies in England and specifically the work with NHS England and the 6 ICSs in the South East is focused on understanding urgent care flow through an acute hospital and flow through the system, which requires admitted patient care data. 3. Work with Health economies in England and specifically NHS England and the work with the 6 ICSs in the South East is looking at system demand which requires Accident and Emergency and Outpatient Data sets Justification for the requested number of years of data: Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields that are relevant to the scope of the projects Lightfoot are engaged in. [1 paragraph unchanged] Clients are only approved if they are an NHS organisation or academic [17 words unchanged] data for the provision of health services or promotion of health. The group ceased when Lightfoot stopped receiving new live data. Upon commencement of this new contract and the reinstatement of refreshed monthly data feeds, the Lightfoot HES Group which consists of representatives from IG, Technical and Production will resume holds monthly meetings to discuss any material changes to the data and to [37 words unchanged] minutes/meeting procedure and the terms of reference as part of this agreement. In addition, the data will not be used for sales or marketing purposes or in compiling tender responses. Lightfoot Solutions are the Data Controller who also processes the data. Lightfootare permitted to process the data to provide the outputs detailed below. NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. • The clients Ligtfoot work with are given direct access to perform analysis on the Hospital Episode Statistics (HES) data using a Business Intelligence platform written and hosted by Lightfoot. The South East Region of the NHS covers a highly diverse population with many factors impacting demand for urgent care services which are now further complicated by the impact of Covid-19. There is significant underlying variation in demand between systems as well as high levels of variation within each system. COVID-19 introduces additional uncertainty for future planning, particularly in relation to Winter 2020/2021. The level of demand driven by patient cohorts at risk of requiring acute care for respiratory related conditions during ‘normal’ winter pressures will have changed significantly in some regions due to the current COVID-19 pandemic. • The results of the analysis are aggregated and automatically have small numbers suppressed before being returned by the application in graph and chart formats Lightfoot’s proposed solution is based on the principal that the underlying growth rate and the complexity of the demand pressures within the region are quantifiable using standard national data sources (HES). This approach can be used to identify areas where variation is not within a normal range to highlight the systems with exceptional levels of demand growth. • Lightfoot classifies the data as a statistic as clients do not have access to record level data. Lightfoot’s proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. • The client determines the nature of any analysis and the final use and audience of that analysis – excepting that the result will be an aggregated statistic. The sfn analytics engine is capable of projecting a cyclical trend with a step change process break, the software currently has this capability which will be configured to account for the step change in demand for urgent care services during peak COVID-19 demand and the return to a ‘new normal’ level for winter 2020. It is based on the approach developed in Christchurch New Zealand after the 2011 earthquake. It enables multiple inflection points to be configured to account for changes in demand for different populations in the region at different points in time. It will produce a dynamic demand prediction which is customizable for individual systems that updates as new data becomes available. Lightfoot provide the software, platform, storage, and access to the data. In all HES customer use cases using statistical analysis, the drill down platform will automatically suppress small numbers before being presented to the user. The user will be informed that some data has been restricted due to small numbers. The user will have the ability to dynamically define a ‘new normal’ level and calculate the trajectory to this point. The analytics engine will then project the data using the underlying cyclic pattern to predict activity levels for urgent care services in winter 2020. Lightfoot are currently working with NHS England and Mid Yorks NHS Hospitals Trust. The output will be a user configurable online interactive dashboard (with small number suppression applied). The dashboard will quantify the projected activity level for population groups and non-elective services over a 12-month period. There will be drill down options, including filtering the data for different time profiles, down to a daily activity prediction. Lightfoot’s clients past and present are either NHS organisations i.e. Providers, Commissioners, ISCs, NHS E, NHS Professional Associations and Academic Health Science Networks. The use case is primarily to support winter planning for 2020, although it may have applications for quantifying the unmet demand resulting from COVID-19. This could include using the tool to quantify ‘difference to expected’ activity level during high levels of demand for COVID patients. Although this is not the primary application, researchers or planners could use the platform to identify the volume and nature of expected activity which never happened in an acute setting which could be described as ‘unmet need’ in the community. The pseudonymised Programme Budgeting Category (PBC) has been requested to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. All organisations party to this agreement must comply with the Data Sharing Framework Contract, including requirements on the use (and purposes of that use) by "Personnel" (as defined within the Data Sharing Framework Contract i.e.: employees, agents and contractors of the Data Recipient who may have access to that data).

Processing activities

All processing of record level data will take place on hardware and [14 words unchanged] within the datacentre without data being transferred to or processed using laptops, desktops desktops, or networks outside of the datacentre. C4L (IDE Group Ltd) Cloud CoCo hosts Lightfoot’s secure rack, servers, power and internet connection within their facility. Cloud CoCo has no access to the data whatsoever. All access by third parties is through the signalsfromnoise (sfn) tool. BI platform. The connections to this tool platform are encrypted over Secure Sockets Layer (SSL) and require the end user [18 words unchanged] charts and tabulations prior to returning the results to the user/third party. [1 paragraph unchanged] i) Psuedonymised i. Pseudonymised HES Data will be downloaded via Secure Electronic File Transfer (SEFT) to [98 words unchanged] with trusted national reference sources it is promoted to the production server. ii) ii. Third parties as specified in this application will only access the data [62 words unchanged] of process performance to provide greater understanding of patient flow and pathways. The sfn tool BI platform provides unique views of the data with functionality (not available in traditional [19 words unchanged] of process performance and provide greater understanding of patient flow and pathways. [2 paragraphs unchanged] The Lightfoot HES group over sees the governance for approving and on-boarding new clients with access to an SFN a sfn platform containing HES data. This group is accountable for ensuring new clients' [21 words unchanged] approved if they are an NHS organisation, NHS health provider or a not for profit not-for-profit academic organisations conducting health research using the data for the provision of health services or promotion of health. Justification for the requested data sets: In addition, the data will not be used for sales or marketing purposes. 1. Ambulance data is required as Lightfoot are looking at the whole patient journey to show outcomes of those patients who arrive via an ambulance. 2. Lightfoot's work with Pen Chord and Exeter University is focused on understanding urgent care flow through an acute hospital and flow through the system, which requires admitted patient care data. 3. Work with NHS England and NHS Improvement is looking at system demand which requires Accident and Emergency and Outpatient Data sets Justification for the requested number of years of data: Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields that are relevant to the scope of the projects Lightfoot are engaged in. [4 paragraphs unchanged] The process for access control for all users of the tool sfn platform is: [3 paragraphs unchanged] • These user requests or User lists need to be validated by [10 words unchanged] through the helpdesk@Lightfootsolutions.com, with a cc to the lead analyst for that particular client. client • All External User requests must be authorised by the nominated client lead and also the Lightfoot Account Manager. Manager [4 paragraphs unchanged] • Once the user has been issued their username and training/confirmation has taken place, the CRM case can be closed closed, and user list updated. [5 paragraphs unchanged] NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. Lightfoot’s proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. The scope of the dashboard goes beyond other ‘peer group’ benchmarking tools to adjust for seasonal pressure, demographic change and the changing health profile of the population. The objective of adjusting for these known explanatory variables (known as standardised benchmarking) is to highlight the systems where these factors do not explain the demand pressure. This feature will enable users to focus on the systems where demand growth is exceptional, in context of its population. Powered by signalsfromnoise, Lightfoot’s propriety Statistical Process Control (SPC) analytics engine, the dashboard will enable users to drill down into sub-populations (to GP practice level) and into specific cohorts in order to understand the correlation between changes in the health profile of the population and the demand pressures. This feature will highlight the specific sub-populations where a high growth rate may be driven by other systemic factors or a behavioural change in the way services are accessed.

Expected output

Lightfoot’s offering is designed to support and enable continuous improvement projects in the NHS and allied health care organisations to improve patient outcomes. Specific outputs include: • Charts and graphical representations of data using statistical process control to highlight variation; Specific outputs include: • Tabulations and summarised data; a) Charts and graphical representations of data using statistical process control to highlight variation • Statistical analysis; b) Tabulations and summarised data • Written reports and recommendations for stakeholders based on findings from analysis and supported by Statistical Process Control (SPC) charts from the sfn tool; c) Statistical analysis • SPC signals and alerts indicating processes where behaviour has recently changed. d) Written reports and recommendations for stakeholders based on findings from analysis and supported by Statistical Process Control (SPC) charts from the sfn tool In all cases these are for Lightfoot’s clients to use with operational staff and clinicians to support service improvement work by presenting HES data in a unique and highly visual manner through the sfn tool. The data is presented in dashboards, charts and in written reports as required by clients. In addition a client may require use of sfn SPC charts to demonstrate to commissioners and other NHS organisations or members of the public where improvements to service levels or patient experience have resulted from their initiatives. In these cases the results analysed through sfn are used in reports and client case studies. All charts and reports included in the outputs use aggregate data (small numbers suppressed) in line with the HES Analysis Guide, derived from the HES data and presented via the sfn tool. e) SPC signals and alerts indicating processes where behaviour has recently changed. Demand growth benchmarking and Recovery The product enables systems to compare the relative urgent care demand growth within their population to identify areas of high growth. Users can drill down into populations (to GP practice level) and understand whether changes in the demographic profile of the population correlate with demand pressure on the urgent care system. The interface is designed to help systems identify areas where demographic change does not explain high levels of demand growth. The project has highlighted specific populations where demand growth was high and proactive intervention could potentially avoid hospitalisation, examples were highlighted in both: • Frimley ICS • Sussex Health and Care Partnership Supporting winter planning The product includes a user configurable demand model to help systems predict acute urgent care activity and occupancy levels to support winter planning for 2022/23. This model accounts for normal seasonal demand pressures and enables users to configure scenarios based on a ‘return to new normal’ following the impact of COVID. Benefits could be shown using demographics and non-demographics to highlight pre COVID demand and growth summary. How good data enabled a GP Practice in Kent to develop, measure and accelerate the delivery of improvement using sfn Lightfoot. sfn helped: • Understand our baseline patient needs and problem we are trying to fix • Understand the improvements we were making and ensure our outcomes were delivering value • Benchmark against neighbouring PCN and the wider Kent system. • Demonstrate the evidence of improvement ‘Making Data Count’ to colleagues utilising quantitative measures Benefits Some early feedback from clinicians, both in primary and secondary care: • Patients, families and carers spend more time in their place of choice – preferably home • Health and Social Care workforce has more time to address the right individuals • Increased ambulance capacity not conveying those who could be cared for at home (or in nursing home) • Better system flow creates capacity to address the medium to long term Elective Care challenge • Significant financial benefits (avoidance/reduction) through beds being occupied by those patients who most require acute care • Creates transformation fund to further improve and expand primary and community services to replicate success across the wider East Kent health and social care system • Improve system safety through professional teams having quality time to proactively measure and monitor patients • Innovative and high performing systems attract the additional workforce required to address the current / future challenge • Improves morale across the health and social care system • Accelerates collaborative working In all cases these are for Lightfoot’s clients to use with operational staff and clinicians to support service improvement work by presenting HES data in a unique and highly visual manner through the sfn tool. The data is presented in dashboards, charts and in written reports as required by clients. In addition, a client may require use of sfn SPC charts to demonstrate to commissioners and other NHS organisations or members of the public where improvements to service levels or patient experience have resulted from their initiatives. In these cases, the results analysed through sfn are used in reports and client case studies. All charts and reports included in the outputs use aggregated data (small numbers suppressed) in line with the HES Analysis Guide, derived from the HES data and presented via the sfn tool. [1 paragraph unchanged]

Expected measurable benefits

NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. Public health decision making continues to become increasingly complex, and the use of data has become essential to this process. To provide the ability for Healthcare Organisations to view their data through a different lens, the implementation of Integrated Care Systems has enhanced the need for a consistent approach to support integration, evaluation, planning and resource investment. By supporting this approach, we can identify opportunities that will improve the efficiency and effectiveness of a health system at a local level to monitor population health and to target interventions. At a national level, this data can be used for effective resource allocation and short-, medium- and long-term planning. These wider public health benefits can include reduced wait times for treatment, reduced admissions to hospital, reduced length of stay upon admission to hospital and improved aftercare. This approach will also benefit the wider public as viewing data in this way will enable faster responses to new threats to public health, such as a reoccurrence of the COVID-19 pandemic. The South East Region of the NHS covers a highly diverse population with many factors impacting demand for urgent care services which are now further complicated by the impact of Covid-19. There is significant underlying variation in demand between systems as well as high levels of variation within each system. COVID-19 introduces additional uncertainty for future planning, particularly in relation to Winter 2020/2021. The level of demand driven by patient cohorts at risk of requiring acute care for respiratory related conditions during ‘normal’ winter pressures will have changed significantly in some regions due to the current COVID-19 pandemic. The high-level benefits for the public, and their health listed both above, and those detailed below are measured through outcomes tracked through the sfn Viewer. Benefits to our clients is the ability to identify, from processing the HES data using sfn, opportunities to reduce bed days as part of winter preparedness and ongoing system planning with a focus on maintaining Elective recovery since the COVID-19 pandemic. Lightfoot's proposed solution is based on the principal that the underlying growth rate and the complexity of the demand pressures within the region are quantifiable using standard national data sources (HES). This approach can be used to identify areas where variation is not within a normal range to highlight the systems with exceptional levels of demand growth. The workstreams are ongoing and use the monthly updates of HES data to monitor improvement and explore changes in variation. Target dates are ongoing as the work extends through the approaching Winter. Our work is focused on identifying key constraints, variation, and bottlenecks to improve the flow of a healthcare system. Lightfoot's proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. Lightfoot is working with a number of Health Economies in England to use HES to identify opportunities to reduce bed days as part of winter preparedness and support ongoing long-term system planning to focus on reducing, the elective backlog. The scope of the dashboard goes beyond other ‘peer group’ benchmarking tools to adjust for seasonal pressure, demographic change and the changing health profile of the population. The objective of adjusting for these known explanatory variables (known as standardised benchmarking) is to highlight the systems where these factors do not explain the demand pressure. This feature will enable users to focus on the systems where demand growth is exceptional, in context of its population. sfn provides an opportunity to view planned care alongside urgent and emergency care (UEC) in a single view, overlaid with our Population Cohort analytics to segment patients and identify which cohorts are having the biggest impact on the system, thus supporting the healthcare provisions on the public. Powered by signalsfromnoise, Lightfoot’s propriety SPC analytics engine, the dashboard will enable users to drill down into sub-populations (to GP practice level) and into specific cohorts in order to understand the correlation between changes in the health profile of the population and the demand pressures. This feature will highlight the specific sub-populations where a high growth rate may be driven by other systemic factors or a behavioural change in the way services are accessed. Using sfn for “what if analysis” allows modelling of the future to test theories and potential scenarios across the health system in live drilldown workshops, providing management and clinical leadership with options for taking different interventions in a sequenced approach that recognises clinical risk. The sfn platform provides the analytics to support business case development and evaluation of the interventions on the fly. The sfn analytics engine is capable of projecting a cyclical trend with a step change process break, the software currently has this capability which will configured to account for the step change in demand for urgent care services during peak COVID-19 demand and the return to a ‘new normal’ level for winter 2020. It is based on the approach developed in Christchurch New Zealand after the 2011 earthquake. It enables multiple inflection points to be configured to account for changes in demand for different populations in the region at different points in time. It will produce a dynamic demand prediction which is customizable for individual systems that updates as new data becomes available. The user will have the ability to dynamically define a ‘new normal’ level and calculate the trajectory to this point. The analytics engine will then project the data using the underlying cyclic pattern to predict activity levels for urgent care services in winter 2020. The work will quantify the demand from patients accessing the UEC pathway from outside catchment areas. sfn capability will also allow users size virtual ward demand – highlighting which cohorts would benefit from a virtual ward intervention and therefore the potential impact on bed occupancy. The output will be a user configurable online interactive dashboard (with small number suppression applied). The dashboard will quantify the projected activity level for population groups and non-elective services over a 12-month period. There will be drill down options, including filtering the data for different time profiles, down to a daily activity prediction. The analysis is typically delivered through a series of data-led workstreams that utilise an analytical capability in sfn which offers a unique view of activity and outcomes by measuring variation at every step of the patient journey. The use case is primarily to support winter planning for 2020, although it may have applications for quantifying the unmet demand resulting from COVID-19. This could include using the tool to quantify ‘difference to expected’ activity level during high levels of demand for COVID patients. Although this is not the primary application, researchers or planners could use the platform to identify the volume and nature of expected activity which never happened in an acute setting which could be described as ‘unmet need’ in the community. Typical workstream focus is on: Lightfoot are planning workshops with senior leadership in East Kent to review the incidents of fractured neck of femur across their CCGs and explore using acute data as part of a group of metrics to measure outcomes for their elderly population to monitor initiatives to be rolled out in line with NICE guidance for frail patients and patients with long term conditions. The HES data in the sfn platform will provide data on numbers to support these discussions. 1. Planning and modelling capability to support systems to address the potential pressures this winter whilst maintaining Elective Recovery 2. sfn Population Cohort analysis to identify opportunities and optimal solutions to mitigate acute capacity pressures 3. Providing a flow-based approach to understanding UEC & elective demand and show population-based variation at ICS, Place and PCN level Benefit of clients being able to use HES analysed through sfn for winter planning Lightfoot are providing a number of clients with immediate planning support and enable a focus on an identified adult population cohort at high risk of an extended admission. Advanced cohort analysis is a specific feature of sfn that clients are unable to replicate easily with inhouse BI tools. Specifically, the feature enables real time cohort building with no lead time to test the results. This allows clinical and operational users to define a cohort and see how those patients interact with services on a timeseries within a few clicks. This rapid feedback cycles enables non-technical users to refine cohort criteria for both intervention planning and evaluation where other methods could take weeks or months with back and forth. Advanced cohort analysis provides a unique insight into how different patients are moving through the system and where potential demand pressures may arise or where interventions will be most pivotal in preventing bottle necks and maintaining flow, from A&E arrivals to time in theatre to post-surgical occupancy. sfn automatically predict levels using the demand history relative to the population and will therefore have the benefit of being able to also assess unwarranted variation within local systems. The sfn analysis has identified cohorts of patients at high risk of an extended admission based on their clinical coding for frailty and previous LOS. This is a small group per GP practice but there is a high utilisation rate so targeting this group as part of Winter planning could significantly reduce pressure on the urgent care system. NHS England and NHS Improvement – Demand Growth Analysis and Projections The NHS covers a highly diverse population with many factors impacting demand. There is significant underlying variation in demand between systems as well as high levels of variation within each system. Lightfoot's solutions are based on the principal that the underlying growth rate and the complexity of the demand pressures are quantifiable using standard national data sources (HES). This approach can be used to identify areas where variation is not within a normal range to highlight the systems with exceptional levels of demand growth. Lightfoot’s solutions focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The scope of the solution goes beyond other ‘peer group’ benchmarking tools to adjust for seasonal pressure, demographic change, and the changing health profile of the population. The objective of adjusting for these known explanatory variables (known as standardised benchmarking) is to highlight the systems where these factors do not explain the demand pressure. This feature enables users to focus on the systems where demand growth is exceptional, in context of its population. Powered by signalsfromnoise, Lightfoot’s propriety SPC analytics engine, the dashboard will enable users to drill down into sub-populations (to GP practice level) and into specific cohorts to understand the correlation between changes in the health profile of the population and the demand pressures. This feature highlights the specific sub-populations where a high growth rate may be driven by other systemic factors or a behavioural change in the way services are accessed. The sfn analytics engine can project a cyclical trend with a step change process break, the software currently has this capability which will configured to account for the step change in demand for urgent care services during peak COVID-19 demand and the return to a ‘new normal’ level. It is based on the approach developed in Christchurch New Zealand after the 2011 earthquake. It enables multiple inflection points to be configured to account for changes in demand for different populations in the region at different points in time. It will produce a dynamic demand prediction which is customizable for individual systems that updates as new data becomes available. The user will have the ability to dynamically define a ‘new normal’ level and calculate the trajectory to this point. The ‘PBC code’ is needed to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. These features enable sfn to provide long term demand projections which consider demographic and other social economic factors such as changing deprivation levels. Clients have used this capability to support strategic planning in several contexts, for example Cardiff and Vale health board Commissioning Lightfoot to build a long term population based demand model to inform planning assumptions for a potential future hospital build 10 years from now. The model is capable of applying the same demand growth assumptions across multi time frames and date partitions. For example, a 10 year monthly projection can be used to predict demand on an hourly timeframe over the next 3 weeks. This enables a unique mix of strategic and operational planning where services can use the same assumptions to predict peak demand during the week and the overall levels over the next decade. Other Benefits [1 paragraph unchanged] Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields requested are relevant to the scope of the projects Lightfoot are engaged in. Provider A&E data is now submitted via the ECDS sfn measures are mapped across the ECDS standard and the legacy A&E CDS which enables long term trend and variation analysis covering the migration period. This is particularly helpful for strategic population-based analysis where long sample periods improve the accuracy of projections. The ‘PBC code’ is needed to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. The national level data available in HES provides sufficiently large sample sizes for meaningful evaluation of services designed to impact small numbers of patients with relatively rare characteristics or low volume services. The HES data provides a large control group for this type of evaluation which often increases the confidence level of any significant results.

Benefits reported

A board of ambulance medical directors representing all trusts have used the data dashboard and it has supported improvement strategies by looking at the variation in acute trust HES data for arrival by ambulance, re-admissions, % that are admitted and acuity of patients conveyed by ambulance to A&E. The HES data is viewed alongside ambulance data to form partial measures for patient outcomes. Some ambulance trusts have then gone on to apply SPC principles to local data and commission support for service improvement. The HES data has identified where there is an opportunity for an ambulance trusts to increase their “hear and treat” rate so patients are treated and remain in their home with community nurse support. Ambulance trusts are trying to move away from a target driven service to a patient centric service and they are reviewing the data collected to support this approach. The following are examples of benefits achieved through use of the sfn BI platform with HES data for Lightfoot clients. Exeter University have modelled how to maximise access to thrombectomy for acute stoke patients with the configuration of stoke services. Redirecting patients directly to attend a comprehensive stroke centre by accepting a small delay (15-min maximum) in intravenous thrombolysis reduces time to endovascular thrombectomy: 25% of all patients would be redirected from hyper-acute stroke units to a comprehensive stroke centre. This would introduce an average delay in intravenous thrombolysis of 8 min, and an average improvement in time to endovascular thrombectomy of 80 min. The balance of comprehensive stroke centre: hyper-acute stroke unit admissions would change from 24:76 to 49:51. Lightfoot worked with an acute provider and their commissioner clinicians to look at frailty pathways and assess if HES data can be used to understand the impact of new frailty clinics in the community. Existing Client Use The following are examples of benefits achieved through use of the sfn tools with HES data for Lightfoot clients. This is currently difficult to quantify as clients have not been able to access new monthly uploads and therefore have been unable to update benefits and outcomes of previous projects. Where possible dates and examples have been included . It is expected that clients will update work once the monthly updates have been obtained and can update benefits accordingly. 1) South West Academic Health Science Network is continuing the use of sfn tools with HES data to work with NIHR CLAHRC South West Peninsula on behalf of NHS England. They have used HES data within the Lightfoot platform to compare patient outcomes by comparing the Somerset Practice Quality Scheme (SPQS) with the national Quality & Outcomes Framework (QOF). The paper has been published, “An evaluation of the Somerset Practice Quality Scheme” in July 2015. The paper makes recommendations to expand the notion of quality in primary care and provide a way to capture what is happening systemically in a second evaluation of SPQS. The HES data continues to support the roll out of this work. http://www.swahsn.com/wp-content/uploads/2016/06/Evaluation-of-the-Somerset-Practice-Quality-Scheme-July-2015.pdf 2) In 2017 South West Academic network used the HES data to support a southwest acute trust to understand the urgent care flow through their acute hospital. HES within the sfn platform was used for a quick drill down to understand why waiting time in ED had increased. The organisations thought it was changes to demand due to changes to Out of Hours (OOHs) providers. They looked at the flow through the system and were able to isolate changes to flows in the hospital as a root cause. Their A&E delivery board used the information to myth bust and pinpoint areas to reduce patient waiting time using the data. 3) Exeter University supported the South West Cardiovascular Strategic Clinical network (SW SCN) in recommendations to reconfigure existing acute services to establish a network of emergency centres for heart attacks and strokes. The purpose of these centres is to maximise good outcomes through the provision of high-quality specialist services that are resilient and sustainable. HES data was used as part of the modelling to develop a clinical benefit measures to look at the number of patients treated for time and volume sensitive conditions of ST (cardiac output) elevated (Myocardial Infarction (MI) and stroke where time to treatment is a big factor in patient outcomes and maintaining function. The proposal for service re-configuration is going through the governance system and a reconfiguration is expected. 4) PenChord (the Peninsula Collaboration for Health Operational Research and Development) part of SW Peninsula CLAHRC (Collaboration for Leadership in Applied Health Research and Care) used HES data over the past year to: • Help Torbay and Devon Foundation trust with understanding the number of beds needed to obtain a good flow of patients through the acute and rehabilitation unit stroke care pathway. The aim is that patients should not be held up at any phase of the care pathway due to lack of beds in the next phase. The model’s findings have been presented to Torbay and South Devon NHS Foundation Trust in a report describing how the resources available affect the Trust’s ability to meet best-practice targets for stroke care. (http:/ninsula.nihr.ac.uk/research/penchord-torbay-and-south-devon-stroke-care) • Conducted research to look at the number and location of neonatal and childbirth centres in England. The HES data for childbirth was used for modelling. This was funded by a National Institute for Health Research and the findings were published in 2017 • Worked with The Royal Cornwall Hospitals NHS Trust to explore the current Cornwall acute and rehab stroke treatment system, and to determine whether ring-fencing beds in acute hospitals and Rehabilitation Services Units (RSUs) would allow those requiring treatment to have a rehab bed in the location of their choice. The team concluded that current rehab bed availability does not match the home location of patients and ‘Ring-fencing’ stroke beds to ensure a free bed is available at a patients closest hospital 90% of the time, would require acute hospitals to run at ~70% average bed occupancy on their stroke wards and Stroke Rehabilitation Units to run at ~75% average bed occupancy. Initial results from the project have been shared with stakeholders via the local Stroke Partnership Board. The results and implications of the work will be reviewed from a commissioning perspective, and will be further shared with Plymouth colleagues, so that learning can be disseminated across the areas studied. (http://clahrc-peninsula.nihr.ac.uk/research/penchord-cornwall-acute-and-community-stroke-bed-capacity-modelling) 5) Lightfoot worked with an acute provider and their commissioner clinicians to look at frailty pathways and assess if HES data can be used to understand the impact of new frailty clinics in the community. [1 paragraph unchanged] 6) Lightfoot has previously provided benchmarking data form HES to the Association of Ambulance Chief Executives Association of Ambulance Chief Executives (AACE) that has allowed the membership of English ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance to inform national policy. NHS England and ICSs in South East have experienced the following benefits listed below from the sfn analytics platform which has measures and configurations available out of the box plus localised configurations that have been deployed as part of the earlier phases of the work. • Analysis of patient data with ambulance trusts identified regions across the country where ambulance trusts were delivering an enhanced clinical model, which allowed increased rates of See, and Treat therefore significantly reducing the number of “avoidable attendances” to A&E departments of patients transported by ambulance. In one region (re: point 8) this established better outcomes for patients but also provided significant financial savings to the health economy when patients were treated at scene rather than be transported to hospital. Using HES data this way allows AACE to share evidence for best practice. This also helped commissioners to appropriately fund this level of service provided by the ambulance trust. Benefits: • Providing benchmarking data to the AACE has allowed the member ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance. The on-going benefit is that the trusts will be able to highlight opportunities for knowledge share and the transfer of best practice to improve patient outcomes in regions that had the greatest variation. • Central to moving towards a patient centric model of healthcare, based around integrated healthcare systems, the ambition is to: 7) Demand growth benchmarking and Recovery • Enable much greater collaborative healthcare design and delivery The product enables systems to compare the relative urgent care demand growth within their population to identify areas of high growth. • To increase the ‘transparency’ of organisational boundaries which may be inhibiting collaboration and consequently patient centered improvement Users can drill down into populations (to GP practice level) and understand whether changes in the demographic profile of the population correlate with demand pressure on the urgent care system. • A key enabler to this strategic change is aligned data and a strong informatics capability, which is centered on people not organisations - for whole populations; cohorts of that population; the pathway experience of individual patients and their outcomes – and driving data more readily into the hands of key decision makers, planners and most importantly our clinicians The interface is designed to help systems identify areas where demographic change does not explain high levels of demand growth. • Analysing and understanding current and past data has value, but being able to forecast forwards in a statistically robust way will support us to make optimal use of the constrained resources across and within systems, optimising outcomes for our populations The project highlighted specific populations where demand growth was high and proactive intervention could potentially avoid hospitalisation, examples were highlighted in: • For systems this would represent a notable shift in how we seek to understand the operation of our healthcare services and the assurance of quality care that follows • Frimley ICS: • An ability to ‘forecast, plan and adjust’ dynamically based on the latest data and insights, will support us to act collectively ‘quickly and decisively’ as we move out of the pandemic and continue our ‘Recovery’ journey • Sussex Health and Care Partnership: • A system to use actionable insights to, agree, implement and evaluate Recovery focused improvements. 8) Supporting winter planning In September 2021 Lightfoot was engaged by the Southeast Region of NHS England to use the HES dataset to provide insights into the effect of the Covid pandemic on the delivery of both Urgent and Emergency Care and Planned Care and to identify the factors affecting both the demand for hospital-based care and the pressure on bed capacity and to identify the interventions that would have the most impact in helping to mitigate UEC demand, The product includes a user configurable demand model to help systems predict acute urgent care activity and occupancy levels to support winter planning for 2020/21. The insights from the analysis were shared with the 6 ICSs and the 18 Acute Trusts across the South-East Region to support their winter readiness planning. This analysis showed that the growth in the demand for urgent care in the post pandemic period was in most cases in line with or below the pre-pandemic trend level, but that there were particular groups of patients who were placing unusual pressures on different systems. The key finding of this analysis was the identification of a cohort of ‘High Risk’ patients who are aged over 50 yrs and have previously had a 14-day length of stay. These patients account for around 2 per cent of the registered GP population in each of the ICSs in the South-East, but account for more than 50% of the occupied beds in the region. This insight has enabled local systems to focus their out of hospital admission avoidance planning and resources on this group of patients with clear benefits in terms of a reduction in the rate at which these patients are being readmitted and a corresponding reduction in the beds occupied by this cohort of patients. Examples where this approach has been adopted to good effect include an admission avoidance program undertaken in Banbury by the Oxford region of the BOB ICB and in East Kent in the Kent and Medway ICB. In addition, Lightfoot has also used the HES data to support the Clinical Leader of the Year for 2021, to evidence the effectiveness of the model of care for older people that he has developed at his practice in Thanet. The success of Lightfoot’s approach in delivering benefits from the use of the HES data has been recognised by the leadership team in the South-East Region who have recently extended Lightfoot’s contract to support the ICBs in the region in recovering from the effects of the pandemic . This model accounts for normal seasonal demand pressures and enables users to configure scenarios based on a ‘return to new normal’ following the impact of COVID. The South-East region has been working towards improving their access to BI resources with our help. By leveraging the robustness of Statistical Process Control (SPC), we have been able to provide the South-East region with powerful insights into the behaviour of real-world processes and patient pathways. An always-on visual control & monitoring system, it has provided robust evidence to support decisions about change or improvement, identifying hidden insights from our processes. They have been able to measure and monitor the performance of any combination of processes from multiple organisations along the full-service pathway. Our dashboards have delivered up-to-the-minute information and reports to people across the organisation. In this way, process issues can be quickly identified, and targeted actions can be decided upon to improve outcomes. With sfn, the visual modelling of key aspects has been achieved in a variety of ways, such as : Annual, monthly, weekly and daily cyclicity of data / Trends and projections / Anomalies and unusual behaviour. This is allowing UK health organisations to compare their performance and outcomes with other similar UK health organisations to improve patient outcomes. Benefits could be shown using demographics and non-demographics to highlight pre COVID demand and growth summary. Using the data provided, NHS England have been provided with the tools to review their modelling structures by being able to consistently review the same data across the South-East Region. By looking deeply into the data, in the viewers we have created they have been able to understand what drivers are causing increased wait times for ambulances. By reviewing the data to understand the differences it has highlighted Portsmouth as delivering differently to other areas of the South-East. A hospital base process is now being initiated to focus on the problem. This will have a wider impact for the communities involved and supports the focus on improving patient outcomes for the benefit of the public. Benefits from Winter Planning initiatives sfn Population Cohort analysis to identify opportunities and optimal solutions to mitigate acute capacity pressures • Identified patients at high risk of an extended admission based on their previous in-hospital behaviour and LOS o This is a small group per GP practice but there is a high utilisation rate so targeting this group as part of winter planning could significantly reduce pressure on the urgent care system • Empirical risk stratification to support high impact interventions, such as the use of virtual wards or high-risk adult cohort interventions • Advanced cohort analysis provided a unique insight into how different patient groups move through the system, where potential demand pressures may arise and where interventions will be most pivotal in preventing queues and maintaining flow • sfn automatically predicted levels using the demand history relative to the population and gave the benefit of being able to also assess unwarranted variation within local systems • Provide analysis and feedback for current models/improvements for community step up and front door services for high-risk adults • sfn was used to test and develop intervention logic and allows users to evaluate, do future modelling and track the success of interventions • sfn is supporting development of the Integrated Care strategy by identifying the cohorts of older, at-risk patients who will benefit most from a more focused approach to personalised care planning and community-based care delivery working. • Identification of the High-Risk cohort - a group of patients, who use 30% to 60% of the beds Providing a flow based approach to understanding UEC & elective demand and show population-based variation at ICS, Place and PCN level • sfn’s population health analytics used to identify opportunities and optimal solutions to mitigate acute capacity pressures • Providing a flow based approach to understanding UEC & elective demand and ability to show population-based variation at ICS, Place and PCN level. • Provide analytics and support to forecast urgent care demand and highlight current constraints in the system, including identification of patients most suitable for management on an ambulatory pathway • Provide insights with regards to disease specific pathways and models of care to manage existing variation in length of stay, quality indicators and patient outcomes • Development of insights that could support the identification of an optimum predicted date of discharge • Share delivery experiences on improving pathways and care models for high-risk adults to improve flow e.g. Right bed first time, stranded patients, virtual wards, discharge support • Provide sfn Viewers and to monitor unscheduled flow and carry out root cause analysis on issues with flow

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Datasets: + HES-ID to MPS-ID HES Accident and Emergency; + HES-ID to MPS-ID HES Admitted Patient Care; + HES-ID to MPS-ID HES Outpatients

Objective for processing

[35 paragraphs unchanged] AMENDMENT REQUEST The pseudonymised Programme Budgeting Category (PBC) has been requested to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. The pseudonymised Programme Budgeting Category (PBC) has been requested to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platfrom refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. All organisations party to this agreement must comply with the Data Sharing Framework Contract, including requirements on the use (and purposes of that use) by "Personnel" (as defined within the Data Sharing Framework Contract i.e.: employees, agents and contractors of the Data Recipient who may have access to that data).

Expected measurable benefits

New Request of use of Data [12 paragraphs unchanged] AMENDMENT REQUEST [1 paragraph unchanged]

Benefits reported

[17 paragraphs unchanged] 7) Demand growth benchmarking and Recovery The product enables systems to compare the relative urgent care demand growth within their population to identify areas of high growth. Users can drill down into populations (to GP practice level) and understand whether changes in the demographic profile of the population correlate with demand pressure on the urgent care system. The interface is designed to help systems identify areas where demographic change does not explain high levels of demand growth. The project highlighted specific populations where demand growth was high and proactive intervention could potentially avoid hospitalisation, examples were highlighted in: • Frimley ICS: • Sussex Health and Care Partnership: 8) Supporting winter planning The product includes a user configurable demand model to help systems predict acute urgent care activity and occupancy levels to support winter planning for 2020/21. This model accounts for normal seasonal demand pressures and enables users to configure scenarios based on a ‘return to new normal’ following the impact of COVID. Benefits could be shown using demographics and non-demographics to highlight pre COVID demand and growth summary.

Changed only in punctuation, spacing or capitalisation: Processing activities.

Unchanged: Expected output.

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

[1 paragraph unchanged] Lightfoot are an organisation who work to help healthcare organisations transition from [7 words unchanged] flow-based system-wide management approach. By incorporating data from different healthcare providers, Lightfoot are is able to measure patient outcomes across the whole pathway, linking all of the services in each patient’s journey. [4 paragraphs unchanged] NHS Providers, Commissioners, STPs, NHS EI, Medical Schools, NHS Professional Associations and Academic Health Science Networks. [2 paragraphs unchanged] 1. The clients are given direct access to perform analysis on the HES Hospital Episode Statistics (HES) data using a Business Intelligence platform written and hosted by Lightfoot. The client can access a suite of measures and dimensions to perform their own analysis. [17 paragraphs unchanged] *New Request of use of Data* [8 paragraphs unchanged] AMENDMENT REQUEST The pseudonymised Programme Budgeting Category (PBC) has been requested to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platfrom refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers.

Processing activities

[2 paragraphs unchanged] All access by third parties is through the signalsfromnoise (sfn) tool. The connections to this tool are encrypted over SSL Secure Sockets Layer (SSL) and require the end user to authenticate. The sfn tool has a [11 words unchanged] charts and tabulations prior to returning the results to the user/third party. [1 paragraph unchanged] i) Psuedonymised HES Data will be downloaded via SEFT Secure Electronic File Transfer (SEFT) to Lightfoot’s secure data centre facility. This data will then be processed [30 words unchanged] loaded into a SQL server database in a format suitable for Lightfoot’s OLAP Online Analytical Processing (OLAP) tool – signals from noise (sfn). Once text based data has been [29 words unchanged] with trusted national reference sources it is promoted to the production server. [10 paragraphs unchanged] Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields requested that are relevant to the scope of the projects Lightfoot are engaged in. [20 paragraphs unchanged] *New Request of use of Data* [3 paragraphs unchanged] Powered by signalsfromnoise, Lightfoot’s propriety SPC Statistical Process Control (SPC) analytics engine, the dashboard will enable users to drill down into sub-populations [44 words unchanged] systemic factors or a behavioural change in the way services are accessed.

Expected measurable benefits

[10 paragraphs unchanged] Existing Client Use Lightfoot are planning workshops with senior leadership in East Kent to review the incidents of fractured neck of femur across their CCGs and explore using acute data as part of a group of metrics to measure outcomes for their elderly population to monitor initiatives to be rolled out in line with NICE guidance for frail patients and patients with long term conditions. The HES data in the sfn platform will provide data on numbers to support these discussions. The following are examples of benefits achieved through use of the sfn tools with HES data for Lightfoot clients. This is currently difficult to quantify as clients have not been able to access new monthly uploads and therefore have been unable to update benefits and outcomes of previous projects. Where possible dates and examples have been included . It is expected that clients will update work once the monthly updates have been obtained and can update benefits accordingly. Lightfoot have been commissioned to support several ambulance trusts in analysis of patient journey data with ambulance trusts to review clinical models in order to deliver the local urgent care strategy and develop a model to treat more people at home and refer to local community services where appropriate. Lightfoot is in discussion with these ambulance services to support redesign of their clinical model to improve patient outcomes. 1) South West Academic Health Science Network is continuing the use of sfn tools with HES data to work with NIHR CLAHRC South West Peninsula on behalf of NHS England. They have used HES data within the Lightfoot platform to compare patient outcomes by comparing the Somerset Practice Quality Scheme (SPQS) with the national Quality & Outcomes Framework (QOF). The paper has been published, “An evaluation of the Somerset Practice Quality Scheme” in July 2015. The paper makes recommendations to expand the notion of quality in primary care and provide a way to capture what is happening systemically in a second evaluation of SPQS. The HES data continues to support the roll out of this work. http://www.swahsn.com/wp-content/uploads/2016/06/Evaluation-of-the-Somerset-Practice-Quality-Scheme-July-2015.pdf Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields requested are relevant to the scope of the projects Lightfoot are engaged in. 2) In 2017 South West Academic network used the HES data to support a southwest acute trust to understand the urgent care flow through their acute hospital. HES within the sfn platform was used for a quick drill down to understand why waiting time in ED had increased. The organisations thought it was changes to demand due to changes to Out of Hours (OOHs) providers. They looked at the flow through the system and were able to isolate changes to flows in the hospital as a root cause. Their A&E delivery board used the information to myth bust and pinpoint areas to reduce patient waiting time using the data. AMENDMENT REQUEST 3) Exeter University supported the South West Cardiovascular Strategic Clinical network (SW SCN) in recommendations to reconfigure existing acute services to establish a network of emergency centres for heart attacks and strokes. The purpose of these centres is to maximise good outcomes through the provision of high-quality specialist services that are resilient and sustainable. HES data was used as part of the modelling to develop a clinical benefit measures to look at the number of patients treated for time and volume sensitive conditions of ST (cardiac output) elevated (Myocardial Infarction (MI) and stroke where time to treatment is a big factor in patient outcomes and maintaining function. The proposal for service re-configuration is going through the governance system and a reconfiguration is expected. The ‘PBC code’ is needed to describe patient pathway types consistently with common approaches used across the NHS. Including it in the data enables the researcher to group activity into pathways consistently with NHS standards such as ‘RightCare’. In practice this means that when the sfn platform refers to ‘Respiratory Pathways’ it counts the exact same activity as data products provided by other suppliers. 4) PenChord (the Peninsula Collaboration for Health Operational Research and Development) part of SW Peninsula CLAHRC (Collaboration for Leadership in Applied Health Research and Care) used HES data over the past year to: • Help Torbay and Devon Foundation trust with understanding the number of beds needed to obtain a good flow of patients through the acute and rehabilitation unit stroke care pathway. The aim is that patients should not be held up at any phase of the care pathway due to lack of beds in the next phase. The model’s findings have been presented to Torbay and South Devon NHS Foundation Trust in a report describing how the resources available affect the Trust’s ability to meet best-practice targets for stroke care. (http:/ninsula.nihr.ac.uk/research/penchord-torbay-and-south-devon-stroke-care) • Conducted research to look at the number and location of neonatal and childbirth centres in England. The HES data for childbirth was used for modelling. This was funded by a National Institute for Health Research and the findings were published in 2017 • Worked with The Royal Cornwall Hospitals NHS Trust to explore the current Cornwall acute and rehab stroke treatment system, and to determine whether ring-fencing beds in acute hospitals and Rehabilitation Services Units (RSUs) would allow those requiring treatment to have a rehab bed in the location of their choice. The team concluded that current rehab bed availability does not match the home location of patients and ‘Ring-fencing’ stroke beds to ensure a free bed is available at a patients closest hospital 90% of the time, would require acute hospitals to run at ~70% average bed occupancy on their stroke wards and Stroke Rehabilitation Units to run at ~75% average bed occupancy. Initial results from the project have been shared with stakeholders via the local Stroke Partnership Board. The results and implications of the work will be reviewed from a commissioning perspective, and will be further shared with Plymouth colleagues, so that learning can be disseminated across the areas studied. (http://clahrc-peninsula.nihr.ac.uk/research/penchord-cornwall-acute-and-community-stroke-bed-capacity-modelling) 5) Lightfoot worked with an acute provider and their commissioner clinicians to look at frailty pathways and assess if HES data can be used to understand the impact of new frailty clinics in the community. The Trust have used the HES data to consider definition of outcome measures for their acute frailty pathways and want to explore collecting data on patient pathway using their own data. The HES data helped demonstrate concepts of frequent data as part of a continuous improvement framework that will require a daily data feed. 6) Lightfoot are planning workshops with senior leadership in East Kent to review the incidents of fractured neck of femur across their CCGs and explore using acute data as part of a group of metrics to measure outcomes for their elderly population to monitor initiatives to be rolled out in line with NICE guidance for frail patients and patients with long term conditions. The HES data in the sfn platform will provide data on numbers to support these discussions. 7) Lightfoot have been commissioned to support several ambulance trusts in analysis of patient journey data with ambulance trusts to review clinical models in order to deliver the local urgent care strategy and develop a model to treat more people at home and refer to local community services where appropriate. Lightfoot is in discussion with these ambulance services to support redesign of their clinical model to improve patient outcomes 8) Lightfoot has previously provided benchmarking data form HES to the Association of Ambulance Chief Executives Association of Ambulance Chief Executives (AACE) that has allowed the membership of English ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance to inform national policy. • Analysis of patient data with ambulance trusts identified regions across the country where ambulance trusts were delivering an enhanced clinical model, which allowed increased rates of See, and Treat therefore significantly reducing the number of “avoidable attendances” to A&E departments of patients transported by ambulance. In one region (re: point 8) this established better outcomes for patients but also provided significant financial savings to the health economy when patients were treated at scene rather than be transported to hospital. Using HES data this way allows AACE to share evidence for best practice. This also helped commissioners to appropriately fund this level of service provided by the ambulance trust. • Providing benchmarking data to the AACE has allowed the member ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance. The on-going benefit is that the trusts will be able to highlight opportunities for knowledge share and the transfer of best practice to improve patient outcomes in regions that had the greatest variation. Based on previous use of HES data over the years, Lightfoot are are only requesting the required data fields requested are relevant to the scope of the projects Lightfoot are engaged in.

Benefits reported

[1 paragraph unchanged] Lightfoot supported AACES’s national objective of improved benchmarking between 10 Ambulance Trusts in England. The yielded benefits included using the HES data as areas of good clinical practice within the 10 trusts and to provide comparison KPI’s. This was then used to roll out improved health outcomes for the regional and national populations. [1 paragraph unchanged] Existing Client Use The following are examples of benefits achieved through use of the sfn tools with HES data for Lightfoot clients. This is currently difficult to quantify as clients have not been able to access new monthly uploads and therefore have been unable to update benefits and outcomes of previous projects. Where possible dates and examples have been included . It is expected that clients will update work once the monthly updates have been obtained and can update benefits accordingly. 1) South West Academic Health Science Network is continuing the use of sfn tools with HES data to work with NIHR CLAHRC South West Peninsula on behalf of NHS England. They have used HES data within the Lightfoot platform to compare patient outcomes by comparing the Somerset Practice Quality Scheme (SPQS) with the national Quality & Outcomes Framework (QOF). The paper has been published, “An evaluation of the Somerset Practice Quality Scheme” in July 2015. The paper makes recommendations to expand the notion of quality in primary care and provide a way to capture what is happening systemically in a second evaluation of SPQS. The HES data continues to support the roll out of this work. http://www.swahsn.com/wp-content/uploads/2016/06/Evaluation-of-the-Somerset-Practice-Quality-Scheme-July-2015.pdf 2) In 2017 South West Academic network used the HES data to support a southwest acute trust to understand the urgent care flow through their acute hospital. HES within the sfn platform was used for a quick drill down to understand why waiting time in ED had increased. The organisations thought it was changes to demand due to changes to Out of Hours (OOHs) providers. They looked at the flow through the system and were able to isolate changes to flows in the hospital as a root cause. Their A&E delivery board used the information to myth bust and pinpoint areas to reduce patient waiting time using the data. 3) Exeter University supported the South West Cardiovascular Strategic Clinical network (SW SCN) in recommendations to reconfigure existing acute services to establish a network of emergency centres for heart attacks and strokes. The purpose of these centres is to maximise good outcomes through the provision of high-quality specialist services that are resilient and sustainable. HES data was used as part of the modelling to develop a clinical benefit measures to look at the number of patients treated for time and volume sensitive conditions of ST (cardiac output) elevated (Myocardial Infarction (MI) and stroke where time to treatment is a big factor in patient outcomes and maintaining function. The proposal for service re-configuration is going through the governance system and a reconfiguration is expected. 4) PenChord (the Peninsula Collaboration for Health Operational Research and Development) part of SW Peninsula CLAHRC (Collaboration for Leadership in Applied Health Research and Care) used HES data over the past year to: • Help Torbay and Devon Foundation trust with understanding the number of beds needed to obtain a good flow of patients through the acute and rehabilitation unit stroke care pathway. The aim is that patients should not be held up at any phase of the care pathway due to lack of beds in the next phase. The model’s findings have been presented to Torbay and South Devon NHS Foundation Trust in a report describing how the resources available affect the Trust’s ability to meet best-practice targets for stroke care. (http:/ninsula.nihr.ac.uk/research/penchord-torbay-and-south-devon-stroke-care) • Conducted research to look at the number and location of neonatal and childbirth centres in England. The HES data for childbirth was used for modelling. This was funded by a National Institute for Health Research and the findings were published in 2017 • Worked with The Royal Cornwall Hospitals NHS Trust to explore the current Cornwall acute and rehab stroke treatment system, and to determine whether ring-fencing beds in acute hospitals and Rehabilitation Services Units (RSUs) would allow those requiring treatment to have a rehab bed in the location of their choice. The team concluded that current rehab bed availability does not match the home location of patients and ‘Ring-fencing’ stroke beds to ensure a free bed is available at a patients closest hospital 90% of the time, would require acute hospitals to run at ~70% average bed occupancy on their stroke wards and Stroke Rehabilitation Units to run at ~75% average bed occupancy. Initial results from the project have been shared with stakeholders via the local Stroke Partnership Board. The results and implications of the work will be reviewed from a commissioning perspective, and will be further shared with Plymouth colleagues, so that learning can be disseminated across the areas studied. (http://clahrc-peninsula.nihr.ac.uk/research/penchord-cornwall-acute-and-community-stroke-bed-capacity-modelling) 5) Lightfoot worked with an acute provider and their commissioner clinicians to look at frailty pathways and assess if HES data can be used to understand the impact of new frailty clinics in the community. The Trust have used the HES data to consider definition of outcome measures for their acute frailty pathways and want to explore collecting data on patient pathway using their own data. The HES data helped demonstrate concepts of frequent data as part of a continuous improvement framework that will require a daily data feed. 6) Lightfoot has previously provided benchmarking data form HES to the Association of Ambulance Chief Executives Association of Ambulance Chief Executives (AACE) that has allowed the membership of English ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance to inform national policy. • Analysis of patient data with ambulance trusts identified regions across the country where ambulance trusts were delivering an enhanced clinical model, which allowed increased rates of See, and Treat therefore significantly reducing the number of “avoidable attendances” to A&E departments of patients transported by ambulance. In one region (re: point 8) this established better outcomes for patients but also provided significant financial savings to the health economy when patients were treated at scene rather than be transported to hospital. Using HES data this way allows AACE to share evidence for best practice. This also helped commissioners to appropriately fund this level of service provided by the ambulance trust. • Providing benchmarking data to the AACE has allowed the member ambulance trusts to explore the variation in outcomes for patients transported to hospital by ambulance. The on-going benefit is that the trusts will be able to highlight opportunities for knowledge share and the transfer of best practice to improve patient outcomes in regions that had the greatest variation.

Unchanged: Expected output.

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Datasets: + Emergency Care Data Set (ECDS)

Objective for processing

[1 paragraph unchanged] Lightfoot are an organisation who work to help healthcare organisations transition from [6 words unchanged] a flow-based system-wide management approach. By incorporating data from different healthcare providers, they Lightfoot are able to measure patient outcomes across the whole pathway, linking all of the services in each patient’s journey. The platform uses statistical process control techniques that relies on time series [61 words unchanged] special cause, change to patient pathways or as part of normal seasonal variation? variation. Work streams are provided with a Statistical Process Control (SPC) view of current pathways for patients and to [20 words unchanged] of the improvement actions and advise when necessary intervention should take place. [1 paragraph unchanged] Lightfoot provide the Signals from Noise signalsfromnoise (sfn) statistical tool, which is used by or for (where Lightfoot are providing the service) the following non-commercial organisations: NHS (Providers, Commissioners), Exeter Medical School, NHS Professional Associations, or Academic Health Science Networks (AHSNs). In all cases, data is for use by operational staff and clinicians to support their work by presenting HES data in a unique and highly visual manner through the Signals From Noise (sfn) tool. NHS Providers, Commissioners, STPs, NHS EI, Medical Schools, NHS Professional Associations and Academic Health Science Networks. In all cases, data is for use by operational staff and clinicians to support their work by presenting HES data in a unique and highly visual manner through the sfn tool. [1 paragraph unchanged] 1. The clients are given direct access to perform analysis on the HES data using a BI Business Intelligence platform written and hosted by Lightfoot. The client can access a suite of measures and dimensions to perform their own analysis. [2 paragraphs unchanged] 4. Lightfoot provide the software, platform, storage and access to the data. In all HES customer use cases statistical analysis and the drill down platform will automatically suppress small numbers before being presented to the user. The user will be informed that some data has been [14 words unchanged] provided to any third party customers or Lightfoot internal consultants and analysts. [11 paragraphs unchanged] The Lightfoot HES group oversees the governance for approving and on-boarding new clients. This group is accountable for ensuring new clients clients' requirements are complying with the Health and Social Care Act 2012 as amended by the Care Act 2014. Clients are only approved if they are an NHS organisation or academic organisation alone or working as part of an academic science network (AHSN) conducting health research using the data for the provision of health services or promotion of health. In addition, The group ceased when Lightfoot stopped receiving new live data. Upon commencement of this new contract and the reinstatement of refreshed monthly data feeds, the Lightfoot HES Group which consists of representatives from IG, Technical and Production will resume monthly meetings to discuss any material changes to the data and to document any new uses of the data, Formal minutes will not be used for sales or marketing purposes or in compiling tender responses. maintained. The terms of reference of the Group should be noted on Lightfoot’s website, and Lightfoot will retain minutes and NHS Digital will audit compliance with minutes/meeting procedure and the terms of reference as part of this agreement. In addition, the data will not be used for sales or marketing purposes or in compiling tender responses. *New Request of use of Data* NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. The South East Region of the NHS covers a highly diverse population with many factors impacting demand for urgent care services which are now further complicated by the impact of Covid-19. There is significant underlying variation in demand between systems as well as high levels of variation within each system. COVID-19 introduces additional uncertainty for future planning, particularly in relation to Winter 2020/2021. The level of demand driven by patient cohorts at risk of requiring acute care for respiratory related conditions during ‘normal’ winter pressures will have changed significantly in some regions due to the current COVID-19 pandemic. Lightfoot’s proposed solution is based on the principal that the underlying growth rate and the complexity of the demand pressures within the region are quantifiable using standard national data sources (HES). This approach can be used to identify areas where variation is not within a normal range to highlight the systems with exceptional levels of demand growth. Lightfoot’s proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. The sfn analytics engine is capable of projecting a cyclical trend with a step change process break, the software currently has this capability which will be configured to account for the step change in demand for urgent care services during peak COVID-19 demand and the return to a ‘new normal’ level for winter 2020. It is based on the approach developed in Christchurch New Zealand after the 2011 earthquake. It enables multiple inflection points to be configured to account for changes in demand for different populations in the region at different points in time. It will produce a dynamic demand prediction which is customizable for individual systems that updates as new data becomes available. The user will have the ability to dynamically define a ‘new normal’ level and calculate the trajectory to this point. The analytics engine will then project the data using the underlying cyclic pattern to predict activity levels for urgent care services in winter 2020. The output will be a user configurable online interactive dashboard (with small number suppression applied). The dashboard will quantify the projected activity level for population groups and non-elective services over a 12-month period. There will be drill down options, including filtering the data for different time profiles, down to a daily activity prediction. The use case is primarily to support winter planning for 2020, although it may have applications for quantifying the unmet demand resulting from COVID-19. This could include using the tool to quantify ‘difference to expected’ activity level during high levels of demand for COVID patients. Although this is not the primary application, researchers or planners could use the platform to identify the volume and nature of expected activity which never happened in an acute setting which could be described as ‘unmet need’ in the community.

Processing activities

[1 paragraph unchanged] C4L (IDE Group Ltd) hosts Lightfoot’s secure rack, servers, power and internet connection within their facility. All access by third parties is through the Signals from Noise signalsfromnoise (sfn) tool. The connections to this tool are encrypted over SSL and [22 words unchanged] charts and tabulations prior to returning the results to the user/third party. [1 paragraph unchanged] i) Psuedonymised HES Data will be downloaded via SEFT to Lightfoot’s secure data centre [91 words unchanged] with trusted national reference sources it is promoted to the production server. ii) Third parties as specified in this application will only access the data via the signals for noise (sfn) sfn tool. Lightfoot’s clients benefit from viewing data through the sfn tool because [47 words unchanged] of process performance to provide greater understanding of patient flow and pathways. [3 paragraphs unchanged] The Lightfoot HES group over sees the governance for approving and on-boarding [6 words unchanged] SFN platform containing HES data. This group is accountable for ensuring new clients clients' requirements are complying with the Health and Social Care Act 2012 as [30 words unchanged] the data for the provision of health services or promotion of health. Justification for the requested data sets: 1. Ambulance data is required as Lightfoot are looking at the whole patient journey to show outcomes of those patients who arrive via an ambulance. 2. Lightfoot's work with Pen Chord and Exeter University is focused on understanding urgent care flow through an acute hospital and flow through the system, which requires admitted patient care data. 3. Work with NHS England and NHS Improvement is looking at system demand which requires Accident and Emergency and Outpatient Data sets Justification for the requested number of years of data: Based on previous use of HES data over the years, Lightfoot are only requesting the required data fields requested are relevant to the scope of the projects Lightfoot are engaged in. Prediction and forecasting is integral to the way that sfn undertakes and presents the results of time series analytics to users. The trends and cycles used to provide projections of future values are based on the last 4 years of data (calculated using a least-squares ARIMA model that compares the same period - week, day or hour - in the preceding years). These trends are projected forward using either a linear or lognormal trend line. In addition to shorter-term projections, sfn also supports a range of forecasting requirements for healthcare organisations. In particular, sfn provides the ability to undertake highly granular population-based forecasts through the inclusion of historical (up to 4 years of data) and projected population data in the system. These rates can be constructed at any level of granularity based on the historical trends in the activity rate by population subgroup. The resultant trends are then applied to the forecast population growth for each subgroup and the results aggregated to provide an overall population-based activity forecast for as long a period into the future as is required. Only 5 years of data is held on a rolling basis. This approach provides a significantly more accurate prediction of the pattern of demand at different times of the year than can be achieved with the simpler allocative models that are typically used elsewhere. In addition, the dimensionalised data model within sfn allows these forecasts to be created on an ‘as required’ basis for any activity and for any sub-selection of the population, whether by age, ethnicity, or domicile. [1 paragraph unchanged] The process for access control for all users of the tool is; is: • Before approving or rejecting a request, an email should be sent [10 words unchanged] and level of access. All users are approved by a designated account manger manager nominated by the client site. • A corresponding case will be created in CRM by the helpdesk. This should be one case per person contains containing the name of the person in the subject line. Only ‘bulk’ requests (6+ on a single request) should be added as a single case. [13 paragraphs unchanged] *New Request of use of Data* NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. Lightfoot’s proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. The scope of the dashboard goes beyond other ‘peer group’ benchmarking tools to adjust for seasonal pressure, demographic change and the changing health profile of the population. The objective of adjusting for these known explanatory variables (known as standardised benchmarking) is to highlight the systems where these factors do not explain the demand pressure. This feature will enable users to focus on the systems where demand growth is exceptional, in context of its population. Powered by signalsfromnoise, Lightfoot’s propriety SPC analytics engine, the dashboard will enable users to drill down into sub-populations (to GP practice level) and into specific cohorts in order to understand the correlation between changes in the health profile of the population and the demand pressures. This feature will highlight the specific sub-populations where a high growth rate may be driven by other systemic factors or a behavioural change in the way services are accessed.

Expected measurable benefits

The following are examples of benefits achieved through use of the sfn tools with HES data for Lightfoot clients. The maturity of current change projects makes it is difficult to quantify benefits and provide dates for all projects at the current time. In these cases, a narrative around expected benefits has been provided. New Request of use of Data 1) South West Academic Health Science Network is continuing the use of sfn tools with HES data to work with NIHR CLAHRC South West Peninsula on behalf of NHS England. They have used HES data within the Lightfoot platform to compare patient outcomes by comparing the Somerset Practice Quality Scheme (SPQS) with the national Quality & Outcomes Framework (QOF). The paper has been published, “An evaluation of the Somerset Practice Quality Scheme” in July 2015. The paper makes recommendations to expand the notion of quality in primary care and provide a way to capture what is happening systemically in a second evaluation of SPQS. The HES data continues to support the roll out of this work. NHS England and NHS Improvement – South East Region has requested Lightfoot provide an Acute Urgent Care Demand Growth Dashboard for the South East Region. The South East Region of the NHS covers a highly diverse population with many factors impacting demand for urgent care services which are now further complicated by the impact of Covid-19. There is significant underlying variation in demand between systems as well as high levels of variation within each system. COVID-19 introduces additional uncertainty for future planning, particularly in relation to Winter 2020/2021. The level of demand driven by patient cohorts at risk of requiring acute care for respiratory related conditions during ‘normal’ winter pressures will have changed significantly in some regions due to the current COVID-19 pandemic. Lightfoot's proposed solution is based on the principal that the underlying growth rate and the complexity of the demand pressures within the region are quantifiable using standard national data sources (HES). This approach can be used to identify areas where variation is not within a normal range to highlight the systems with exceptional levels of demand growth. Lightfoot's proposal is to build a dashboard that will identify the systems where the underlying rate of demand growth is higher than predicted. The tool will focus on the changing rate, rather than the absolute value to predict which systems are likely to suffer operational challenges if the projected demand growth is not mitigated. The demand model powering the dashboard will factor in the impact of COVID-19 and can therefore be used to support Winter Planning for 2020/2021 and beyond. The scope of the dashboard goes beyond other ‘peer group’ benchmarking tools to adjust for seasonal pressure, demographic change and the changing health profile of the population. The objective of adjusting for these known explanatory variables (known as standardised benchmarking) is to highlight the systems where these factors do not explain the demand pressure. This feature will enable users to focus on the systems where demand growth is exceptional, in context of its population. Powered by signalsfromnoise, Lightfoot’s propriety SPC analytics engine, the dashboard will enable users to drill down into sub-populations (to GP practice level) and into specific cohorts in order to understand the correlation between changes in the health profile of the population and the demand pressures. This feature will highlight the specific sub-populations where a high growth rate may be driven by other systemic factors or a behavioural change in the way services are accessed. The sfn analytics engine is capable of projecting a cyclical trend with a step change process break, the software currently has this capability which will configured to account for the step change in demand for urgent care services during peak COVID-19 demand and the return to a ‘new normal’ level for winter 2020. It is based on the approach developed in Christchurch New Zealand after the 2011 earthquake. It enables multiple inflection points to be configured to account for changes in demand for different populations in the region at different points in time. It will produce a dynamic demand prediction which is customizable for individual systems that updates as new data becomes available. The user will have the ability to dynamically define a ‘new normal’ level and calculate the trajectory to this point. The analytics engine will then project the data using the underlying cyclic pattern to predict activity levels for urgent care services in winter 2020. The output will be a user configurable online interactive dashboard (with small number suppression applied). The dashboard will quantify the projected activity level for population groups and non-elective services over a 12-month period. There will be drill down options, including filtering the data for different time profiles, down to a daily activity prediction. The use case is primarily to support winter planning for 2020, although it may have applications for quantifying the unmet demand resulting from COVID-19. This could include using the tool to quantify ‘difference to expected’ activity level during high levels of demand for COVID patients. Although this is not the primary application, researchers or planners could use the platform to identify the volume and nature of expected activity which never happened in an acute setting which could be described as ‘unmet need’ in the community. Existing Client Use The following are examples of benefits achieved through use of the sfn tools with HES data for Lightfoot clients. This is currently difficult to quantify as clients have not been able to access new monthly uploads and therefore have been unable to update benefits and outcomes of previous projects. Where possible dates and examples have been included . It is expected that clients will update work once the monthly updates have been obtained and can update benefits accordingly. 1) South West Academic Health Science Network is continuing the use of sfn tools with HES data to work with NIHR CLAHRC South West Peninsula on behalf of NHS England. They have used HES data within the Lightfoot platform to compare patient outcomes by comparing the Somerset Practice Quality Scheme (SPQS) with the national Quality & Outcomes Framework (QOF). The paper has been published, “An evaluation of the Somerset Practice Quality Scheme” in July 2015. The paper makes recommendations to expand the notion of quality in primary care and provide a way to capture what is happening systemically in a second evaluation of SPQS. The HES data continues to support the roll out of this work. http://www.swahsn.com/wp-content/uploads/2016/06/Evaluation-of-the-Somerset-Practice-Quality-Scheme-July-2015.pdf [1 paragraph unchanged] 3) Exeter University supported the South West Cardiovascular Strategic Clinical network (SW [18 words unchanged] and strokes. The purpose of these centres is to maximise good outcomes though through the provision of high quality high-quality specialist services that are resilient and sustainable. HES data was used as [52 words unchanged] re-configuration is going through the governance system and a reconfiguration is expected. 4) Pen chord PenChord (the Peninsula Collaboration for Health Operational Research and Development) part of SW [7 words unchanged] Health Research and Care) used HES data over the past year to: [1 paragraph unchanged] • Conducted research to look at the number and location of neonatal and childbirth centres in England. The HES data for childbirth was used for modelling. This was funded by a National Institute for Health Research and the findings were published in 2017 [4 paragraphs unchanged] 6) Lightfoot are planning a workshop workshops with Greater Manchester senior leadership in East Kent to review the incidence incidents of fractured neck of femur across their CCGs and explore using acute [40 words unchanged] the sfn platform will provide data on numbers to support these discussions. [1 paragraph unchanged] 8) Lightfoot continue to provide has previously provided benchmarking data form HES to the Association of Ambulance Chief Executives Association [19 words unchanged] outcomes for patients transported to hospital by ambulance to inform national policy. • Analysis of patient data with ambulance trusts identified regions across the [25 words unchanged] attendances” to A&E departments of patients transported by ambulance. In one region (re (re: point 8) this established better outcomes for patients but also provided significant financial savings to the health economy when patients were treated at scene rather than be transported to hospital. Using HEs HES data this way allows ACCE AACE to share evidence for best practice. This also helped commissioners to appropriately fund this level of service provided by the ambulance trust. • Providing benchmarking data to the AACE has allowed the member ambulance [13 words unchanged] ambulance. The on-going benefit is that the trusts will be able to highlighted highlight opportunities for knowledge share and the transfer of best practice to improve patient outcomes in regions that had the greatest variation. • Association of Ambulance Chief Executives (AACE): provision of nationwide benchmarking solution to ACCE and ten national ambulance trusts utilising HES data. Supporting strategic objectives in their National Programme. Completing evidence based research to support national commissioning discussions. Based on previous use of HES data over the years, Lightfoot are are only requesting the required data fields requested are relevant to the scope of the projects Lightfoot are engaged in.

Benefits reported

A board of ambulance medical directors representing all trusts use have used the data dashboard and it informs has supported improvement strategies by looking at the variation in acute trust HES data [103 words unchanged] service and they are reviewing the data collected to support this approach. Lightfoot supported AACES’s national objective of improved benchmarking between 10 Ambulance Trusts in England. The yielded benefits included using the HES data as areas of good clinical practice within the 10 trusts and to provide comparison KPI’s. This was then used to roll out improved health outcomes for the regional and national populations. [1 paragraph unchanged]

Unchanged: Expected output.