NHS England - Infections & Antimicrobial Resistance (AMR) Trusted Research Environment - renewal application for the same data fields and datasets and an amendment to the data processor

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

[8 paragraphs unchanged] The purpose of the Antimicrobial Resistance (AMR) AMR programme is therefore to work collaboratively to deliver on the following ambitions set out in the 5 Year NAP: [5 paragraphs unchanged] ● The growing burden on bloodstream infections (BIS) on health services and the need to understand better ways of preventing and managing BIS bloodstream infections so as to improve outcomes and reduce costs [1 paragraph unchanged] In turn, in order to achieve the above ambitions, there is a [17 words unchanged] in due course, made available directly to NHS England (NHSE) so that NHSE NHSE, with support being provided by North Of England Commissioning Support Unit (CSU) - commissioned specifically to deliver the dashboard in NHSE's environment - can develop a dashboard service for NHSE AMR stakeholders who form part of the wider programme and NHS system. North of England CSU do not have access to the NHS Digital patient-level data held in the online portal and thus are not acting as a data processor under this Agreement. Key stakeholders include NHSE central and regional teams, CCGs, clinical commissioning groups, Integrated Care Systems, clinicians (primary and secondary care), providers providers, Department for Health and Public Social Care (DHSC - AMR Function) and UK Health England (PHE). Security Agency (UKHSA). The NHSE Medical Director, is the SRO for the Human Health elements of the NAP and so NHSE has a direct role in implementing the plan and in supporting partner organisations to implement key actions which fall within their responsibilities, including providing data to inform this work. These national, regional and local stakeholders require a common source of information that provides core intelligence regarding infection, antimicrobial prescribing and associated resistance, in order to monitor and identify associated interventions for specific cohorts of patients. The DHSC AMR Function are responsible for the AMR UK agenda and chair the AMR UK Delivery Board which the AMR Programme Board reports to. UKHSA are responsible for the surveillance AMR delivery programme and thus play a role in the development of and the outputs of the dashboard. These stakeholders are not determining the purpose or the means of the processing of NHS Digital data under this Agreement, and are therefore not considered data controllers. These stakeholders do not have access to NHS Digital data held in the online portal and thus are not acting as data processors under this Agreement. The datasets to be made directly available to NHSE are: ● Hospital Episodes Statistics (HES): to identify admissions, diagnoses and other clinical information relating to patients with an infection ● Medicines dispensed in Primary Care (NHSBSA data): to identify key data regarding prescriptions for antibiotics ● Civil Registration Deaths: to identify mortality rates and reasons for mortality ● Community Services Data Set: to address health inequalities by monitoring outcomes ● GP Data for Pandemic Planning and Research: to provide key insights into the impact of the COVID pandemic on AMR ● Electronic Prescribing and Medicines Administration: to provide key insights into the changes of secondary care prescribing during the pandemic By linking and enabling access to the above data, NHSE will have a much richer source of intelligence of patients with infections and use of antimicrobial treatment, so as to support the government’s ambitions in the UK 5 year NAP. The purpose is as follows: • To assess the utility of the data to support the future operating model • To assess the utility of the data to support the testing of hypotheses and discovery of new trends, for example: o relating to cause of deterioration of admitted patients o relating to management of infection along a full patient pathway between different care settings o relating to antibiotic prescribing and its impact on the management of infection The NHSE Medical Director is the Senior Responsible Owner for the Human Health elements of the NAP and so NHSE has a direct role in implementing the plan and in supporting partner organisations to implement key actions which fall within their responsibilities, including providing data to inform this work. A number of workstreams have been set up by NHSE to support delivery of these elements. These workstreams rely on robust data from a plethora of data collections that capture information relating to infection, antimicrobial prescribing and patient outcomes across the healthcare system (i.e. community, primary and secondary care) in order to assess current and historical rates of infection, prescribing rates and prescribing efficacy across the patient pathway, so that workstreams can develop and implement effective plans that reduce drug resistance and ultimately lead to better outcomes for patients. The clinical workstream leads have reviewed the datasets in detail, and identified the fields that they believe will bring value for their individual workstreams and support them to undertake the assessments described. Whilst some minimisation has been applied by restricting the fields to only those needed, further minimisation is anticipated once NHSE analysts have been able to access the data. The purpose of the exploration is to understand exactly which data items can support the clinical workstreams in their aims. Without full sight of the data fields as defined in the supporting spreadsheet, and over the period of time requested, NHSE are unable to evaluate the gaps and thereby the programme is at risk of missing vital information to meet the human healthcare ambitions. The Government’s published UK National Action Plan provides the relevant information in respect of human healthcare (and other) ambitions – UK_AMR_5_year_national_action_plan.pdf (publishing.service.gov.uk). There are no alternative, less intrusive ways of achieving the purpose. Pseudonymised record level data is required in order to link the datasets. The full available time period for each dataset is requested in order to be able to fully assess trends over time at different geographical scales. [1 paragraph unchanged] To do this, NHSE has so far undertaken undertook an initial discovery phase of work by commissioning NHS Digital to design, develop and deliver a Trusted Research Environment (TRE) an online portal by end of March 2021, that enables enabled national datasets relating to AMR to be imported to the TRE, imported; the associated data linked and made available via a dashboard functionality prototype. NHS Digital were a data processor for the discovery phase. From July 2021, the NHS Digital commission came to an end, based on a decision to develop the Infection and AMR Dashboard product within NHS England’s National Data Platform. Thus, NHS Digital are no longer acting as a data processor, and are also not determining the purpose or means of this agreement. the processing. However, system-level access to the datasets identified as integral for understanding patterns of infection and prescribing within NHS Digital’s system remains integral in providing the ability to understanding and explore linkage across key datasets. [1 paragraph unchanged] The intention of a business intelligence infection and AMR dashboard tool in the longer term is primarily to: [4 paragraphs unchanged] The next post-discovery phase requires that NHSE staff, with NHSE as Data Controller, access the tools and data in the TRE online portal to understand and support next steps. This is so that NHSE analysts are able to analyse and manipulate the data in the TRE, so to inform, develop and implement a more permanent operating model during FY 21/22, whereby NHSE can build and develop a reliable stream of intelligence in line with the AMR programme’s needs and in line with the healthcare commitments set out in the 5 year National Action Plan. During FY21-22, NHSE analysts have developed an initial stream of intelligence via an NHSE AMR Dashboard in line with the commitments set out in the 5 year NAP. This application is for the post-discovery phase only, to grant access to the NHS Digital TRE environment, to enable NHSE to gain a sufficient understanding of the data to shape development of the future operating model. During FY22-23, NHSE analysts will continue to analyse and manipulate the data in the online portal to continue to inform, develop and implement a more permanent operating model. No other organisations other than NHSE will access or be involved in processing the data (other than NHS Digital staff who will provide support to the NHSE users). This Agreement is for continued access to the NHS Digital online portal, to enable NHSE to gain a sufficient understanding of the data to shape development of the future operating model (aka the Dashboard product). No other organisations other than NHSE will access or be involved in processing the data. [1 paragraph unchanged] NHS England are processing the data being accessed under this agreement Agreement as part of the public task around research under Article 6(1)(e) and 9(2)(j) of the UK GDPR so as to undertake public health monitoring and statistics, in the public’s interest. [1 paragraph unchanged] It is within the public interest because NHS England are required to [7 words unchanged] the commitments set out in the UK AMR 5 Year National Action Plan, Plan. NHSE need to have reliable, detailed data about people susceptible to infection [37 words unchanged] process of becoming resistant to antibiotics due over prescription/inappropriate usage of antibiotics). [3 paragraphs unchanged] GDPPR data is required for all ages, so as to enable the AMR programme to understand the impact that the COVID pandemic has had on the utilisation of antimicrobial agents and associated resistance changes versus patterns pre-pandemic, taking into account the indications that there have been significant changes in primary prescribing patterns, in order to inform the key actions to ensure effective antimicrobial stewardship and appropriateness of clinical prescribing going forward as the country starts to come out of the pandemic. ages. Without access to GDPRR data relating to children and young people, the programme are unable to understand changes to prescribing patterns for this cohort of patients. AMENDMENT REQUEST This amendment is requested due to the identification of additional fields within HES, Civil Registration (Deaths), CSDS and GDPPR that are integral to the exploratory work within the NHS DAE. The fields requested are all pseudonymous.

Processing activities

The datasets and the reason for requiring access to this national data within the TRE are: NHSE request access to the below listed datasets within NHS Digital’s online portal: ● Hospital Episodes Statistics (HES): to identify admissions, diagnoses and other clinical information relating to patients with an infection ● Hospital Episodes Statistics (HES) Admitted Patient Care, Critical Care, Outpatients, Accident & Emergency / Emergency Care Data Set ● Medicines dispensed in Primary Care (NHSBSA data): to identify key data regarding prescriptions for antibiotics data) ● Civil Registration Deaths: to identify mortality rates and reasons for mortality ● Civil Registration Deaths ● Community Services Data Set: to address health inequalities by monitoring outcomes Set ● GP Data for Pandemic Planning and Research: to provide key insights into the impact of the COVID pandemic on AMR Research ● Electronic Prescribing and Medicines Administration: to provide key insights into the changes of secondary care prescribing during the pandemic ● Electronic Prescribing and Medicines Administration By linking and enabling access to the above data, NHSE will have a much richer source of intelligence of patients with infections and use of antimicrobial treatment, so as to support the government’s ambitions in the UK 5 year National Action Plan (NAP). The purpose is as follows: These datasets will be linkable. There will be no linkage to other data. • To assess the utility of the data to support the future operating model • To assess the utility of the data to support the testing of hypotheses and discovery of new trends, for example: o relating to cause of deterioration of admitted patients o relating to management of infection along a full patient pathway between different care settings o relating to antibiotic prescribing and its impact on the management of infection [1 paragraph unchanged] The data within the TRE will be pseudonymised. The data within the online portal will be pseudonymised. No identifiable data will be accessible within the online portal. There will be no requirement or attempt to re-identify individuals. No identifiable data will be accessible within the TRE. The NHS Digital online portal is hosted and access is audited by NHS Digital to ensure that use is appropriate and in line with the terms agreed in the Data Sharing Agreement. The NHS Digital Portal (TRE) system is hosted and access is audited by NHS Digital to ensure that use is appropriate and in line with the terms agreed in the Data Sharing Agreement. Only NHSE substantive employees are able to access the datasets detailed within this Agreement. Access to the datasets detailed within this Agreement is within the NHS Digital online portal. NHSE substantive employees are only able to access the datasets detailed within this agreement. Access to the datasets detailed within this agreement is within the NHS Digital TRE. All NHSE users will comply with the use of the data as specified in this Agreement. Only registered users will be able to access the online portal. All NHSE users will comply with the use of the data as specified in this agreement. Only registered TRE users will be able to access the TRE. This Agreement is for the post-discovery phase only, to grant access to the NHS Digital online portal, to enable NHSE to gain a sufficient understanding of the data to shape development of the future operating model. Therefore this work is not being carried out in the NHS England COVID-19 data store. EPMA (hospital medicines) data is not currently available within NHSE and thus exploration and analysis of this dataset in conjunction with the other datasets available via the NHS Digital DAE, to support the AMR programme’s needs (spanning multiple datasets), is not yet possible within the NHS England data store. Until such time, access to this dataset alongside all others listed is required via the NHS Digital DAE. This application is purely for the post-discovery phase only, to grant access to the NHS Digital TRE environment, to enable NHSE to gain a sufficient understanding of the data to shape development of the future operating model. Therefore this work is not being carried out in the NHS England COVID-19 datastore. [2 paragraphs unchanged] Registered NHSE employees who are granted access to the TRE, will be supported in their use of the data by NHS Digital staff who have appropriate approved access to the data. The NHSE Medical Director is the SRO for overseeing delivery of the Human Heath elements of the cross-government UK AMR National Action Plan (NAP). A number of workstreams have been set up by NHSE to support delivery of these elements. These workstreams rely on robust data from a plethora of data collections that capture information relating to infection, antimicrobial prescribing and patient outcomes across the healthcare system (i.e. community, primary and secondary care) in order to assess current and historical rates of infection, prescribing rates and prescribing efficacy across the patient pathway, so that workstreams can develop and implement effective plans that reduce drug resistance and ultimately lead to better outcomes for patients. The clinical workstream leads have reviewed the datasets in detail, and identified the fields that they believe will bring value for their individual workstreams and support them to undertake the assessments described. Whilst some minimisation has been applied by restricting the fields to only those needed, further minimisation is anticipated once NHSE analysts have been able to access the data. The purpose of the exploration is to understand exactly which data items can support the clinical workstreams in their aims. Without full sight of the data fields as defined in the supporting spreadsheet, and over the period of time requested, NHSE are unable to evaluate the gaps and thereby the programme is at risk of missing vital information to meet the human healthcare ambitions. The Government’s published UK National Action Plan provides the relevant information in respect of human healthcare (and other) ambitions – UK_AMR_5_year_national_action_plan.pdf (publishing.service.gov.uk).

Expected output

NHS Digital developers will collaborate with NHSE registered users will use the NHS Digital online portal data to produce static support review, evaluation, design and production of visualisations of for the data, AMR Dashboard tool, which will be shared with AMR programme stakeholders in order to demonstrate the continuing value of the work being undertaken. AMR programme stakeholders include clinicians (via clinical commissioning groups, Trusts and integrated care systems), NHS England & Improvement national and regional staff, the UK Health Security Agency, and the Department for Health and Social Care. These static visualisations programme stakeholders will not contain record level data, they will show aggregated data with small numbers supressed. They will support the purpose described above, and contribute to the understanding be ultimate end users of the value of the data, but will not themselves be used as part of a service. dashboard tool. Access to the pseudonymised record level data is expected to provide NHSE with a sufficient level of understanding of the data, to inform the plan for development of a dashboard service in the near future in line with delivering the commitments set out in the 5 year UK AMR National Action Plan (NAP). Visualisations will not contain record level data, they will show aggregated data. It will support the purposes described and contribute to the understanding of the value of the data. The outputs from this initial stage will enable more robust stakeholder engagement in term of developing next steps. Access to the pseudonymised record level data is expected to provide NHSE with a sufficient level of understanding of the data, to inform the plan for development of a dashboard service in line with delivering the commitments set out in the 5 year UK AMR NAP. The outputs will enable more robust stakeholder engagement in term of developing next steps.

Expected measurable benefits

[7 paragraphs unchanged] Through having access to GDPPR and EPMA data, alongside the other linkable datasets (as listed within this agreement only, within the TRE), Agreement only), the AMR programme will be able to identify whether and where patients [9 words unchanged] hospital admission and explore associated prescribing patterns and patient outcomes within hospital. By exploring patients who consult in primary care (using the GDPPR data), if they are prescribed or are not prescribed an antimicrobial (using the Primary Care Prescribing BSA data), end up being admitted to hospital (using the HES data) and [18 words unchanged] clinical outcomes across primary & secondary for different cohorts of patients thereby: [5 paragraphs unchanged] The researcher programme will use the pseudonymised GDPPR COVID-19 data and EPMA COVID-19 data to explore how to provide intelligence to support their response to the COVID-19 emergency and to better understand and plan the impacts on NHS Services and patient outcomes in relation to infection management and AMR. EPMA, GDPPR, HES, Primary care medicines and Civil Registrations Deaths data have immediate value to a number of workstream leads. NHSE have plans to produce novel analysis related to the treatment of specific infections using these datasets over FY22/23. Specific planned work includes development of metrics that identify which antibiotics are most commonly used to treat pneumonia, on what day of admission these are prescribed and how long the antibiotic course was. The datasets also remain invaluable to supporting the development of the longer term operating model. In particular, the ability to link EPMA data to other datasets is enabling novel analysis that will be further developed over FY22/23.

Benefits reported

As access to the data has only been granted in August 2021 and the date of the amendment is August 2021, there are not yet any yielded benefits. Access to NHS Digital's online portal has supported the launch of the initial dashboard in late February 2022, using SUS data held separately by NHSEI as the basis for displaying infection trends and patterns in secondary care alongside mortality rates and lengths of stay at differing regional levels. Metrics have been created to date showing changes in antibiotic course length during the pandemic, overall, and specifically courses of intravenous and oral antibiotics. While the number of courses fell substantially during the early stages of the pandemic, the average course length increase very slightly. The initial dashboard product, whilst only in its infancy, is already proving to be a robust and useful intelligence tool for the AMR programme, and one which will evolve over time to incorporate other core metrics to support the AMR ambition, through understanding and use of other integral datasets available within NHSEI. Thus, system-level access to NHS Digital's online portal is proving useful in evaluating and discussing the needs, priorities and feasibilities of the dashboard product. The analysis done to date using the datasets in the online portal has given NHSE a clear understanding of how linkage to patient-level surveillance datasets could be achieved, which NHSE hope will become available in the coming years.

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

Objective for processing

[34 paragraphs unchanged] AMENDMENT REQUEST The agreement includes GPES Data for Pandemic Planning and Research (COVID-19) (GDPPR) and Electronic Prescribing and Medicines Administration (EPMA) data in Secondary Care for COVID-19. The agreement is to now include GPES Data for Pandemic Planning and Research (COVID-19) (GDPPR) and Electronic Prescribing and Medicines Administration (EPMA) data in Secondary Care for COVID-19. [3 paragraphs unchanged] AMENDMENT REQUEST This amendment is requested due to the identification of additional fields within HES, Civil Registration (Deaths), CSDS and GDPPR that are integral to the exploratory work within the NHS DAE. The fields requested are all pseudonymous.

Expected measurable benefits

[7 paragraphs unchanged] AMENDMENT REQUEST [8 paragraphs unchanged]

Benefits reported

Not stated in the previous version; added here.

As access to the data has only been granted in August 2021 and the date of the amendment is August 2021, there are not yet any yielded benefits.

Unchanged: Processing activities, Expected output.

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

Datasets: + COVID-19 General Practice Extraction Service (GPES) Data for Pandemic Planning and Research (GDPPR); + Electronic Prescribing and Medicines Administration (EPMA) data in Secondary Care for COVID-19

Objective for processing

[7 paragraphs unchanged] More so, there is an emerging threat of pan-antibiotic resistance and equally at the same time slow growth and/or development of new treatments. [26 paragraphs unchanged] AMENDMENT REQUEST The agreement is to now include GPES Data for Pandemic Planning and Research (COVID-19) (GDPPR) and Electronic Prescribing and Medicines Administration (EPMA) data in Secondary Care for COVID-19. Access is requested to enable the AMR programme to understand the impact that the COVID pandemic has had on the utilisation of antimicrobial agents and associated resistance changes versus patterns pre-pandemic, taking into account the indications that there have been significant changes in primary prescribing patterns, in order to inform the key actions to ensure effective antimicrobial stewardship and appropriateness of clinical prescribing going forward as the country starts to come out of the pandemic. This data is required in order to support exploration of trends associated with infection management along a full patient pathway, between different care settings i.e. between primary and secondary care. The application is for system-level access to GDPPR (and EPMA) data for exploratory purposes and thus there will be no published results/outputs and will be in compliance with the permitted use of the GDPPR data collection. GDPPR data is required for all ages, so as to enable the AMR programme to understand the impact that the COVID pandemic has had on the utilisation of antimicrobial agents and associated resistance changes versus patterns pre-pandemic, taking into account the indications that there have been significant changes in primary prescribing patterns, in order to inform the key actions to ensure effective antimicrobial stewardship and appropriateness of clinical prescribing going forward as the country starts to come out of the pandemic. Without access to GDPRR data relating to children and young people, the programme are unable to understand changes to prescribing patterns for this cohort of patients.

Processing activities

[5 paragraphs unchanged] ● GP Data for Pandemic Planning and Research: to provide key insights into the impact of the COVID pandemic on AMR ● Electronic Prescribing and Medicines Administration: to provide key insights into the changes of secondary care prescribing during the pandemic [12 paragraphs unchanged] This application is purely for the post-discovery phase only, to grant access to the NHS Digital TRE environment, to enable NHSE to gain a sufficient understanding of the data to shape development of the future operating model. Therefore this work is not being carried out in the NHS England COVID-19 datastore. [4 paragraphs unchanged] The clinical workstream leads have reviewed the datasets in detail, and identified [100 words unchanged] at risk of missing vital information to meet the human healthcare ambitions. The Government’s published UK National Action Plan provides the relevant information in respect of human healthcare (and other) ambitions – UK_AMR_5_year_national_action_plan.pdf (publishing.service.gov.uk).

Expected measurable benefits

[7 paragraphs unchanged] AMENDMENT REQUEST Through having access to GDPPR and EPMA data, alongside the other linkable datasets (as listed within this agreement only, within the TRE), the AMR programme will be able to identify whether and where patients are failing treatment in primary care, result in a hospital admission and explore associated prescribing patterns and patient outcomes within hospital. By exploring patients who consult in primary care (using the GDPPR data), if they are prescribed or are not prescribed an antimicrobial (using the Primary Care Prescribing BSA data), end up being admitted to hospital (using the HES data) and are prescribed antibiotics within hospital (using the EPMA data), the researcher can better identify health-seeking behaviour, prescribing and clinical outcomes across primary & secondary for different cohorts of patients thereby: • Providing insight into how medicines have been used to treat Covid-19 patients in hospitals, taking into account that Covid is a viral rather than bacterial infection and so the patterns of use of antimicrobials need to be understood • Understanding patterns of prescribing before, during and after Covid-19, taking into account changed cohorts of patients who have presented • Enabling modelling of the impact that treatment of serious infections including antibiotics, antivirals and antifungals has on patient outcomes In turn, helping the programme to develop the right standards to underpin hospital prescribing and develop organisational and healthcare system-wide policies and approaches to promote and monitor judicious use of antimicrobials, to preserve their effectiveness and mitigate the future impact of antimicrobial resistance. Without access to this data, the programme is unable to explore patterns of prescribing within hospital that are key to understanding the patient pathway, nor make sense of patterns of prescribing in line with, for example, comorbidities relevant to infection and medication reviews in primary care that relate to recurrent infection. The researcher will use the pseudonymised GDPPR COVID-19 data and EPMA COVID-19 data to explore how to provide intelligence to support their response to the COVID-19 emergency and to better understand and plan the impacts on NHS Services and patient outcomes in relation to infection management and AMR.

Benefits reported

Stated in the previous version and removed here.

Yielded Benefits is not a requirement for new applications.

Unchanged: Expected output.