NHS Digital Data Release Register - reformatted
NHS North of England Commissioning Support Unit
Project 1 — DARS-NIC-08095-P4D0D
Opt outs honoured: No - data flow is not identifiable
Sensitive: Non Sensitive
When: 2016/04 (or before) — 2019/02.
Legal basis: Health and Social Care Act 2012, Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)
Categories: Anonymised - ICO code compliant
- Hospital Episode Statistics Accident and Emergency
- Hospital Episode Statistics Admitted Patient Care
- Hospital Episode Statistics Critical Care
- Hospital Episode Statistics Outpatients
NHS North of England Commissioning Support (NECS) provides a comprehensive business intelligence (BI) service to a wide range of NHS organisations. This includes both standard analytics and reporting, and deep-dives and diagnostic exercises to offer insight and intelligence on a commissioner’s health economy. In addition NECS offer a mature business intelligence application (RAIDR) allowing self-service access to a range of dashboards and configurable reports. This tool is available on a subscription basis only to NHS organisations and local authorities. It is currently used by Clinical Commissioning Groups (CCGs), internally within the CSU through specialist support teams and by CCG member practices. A list of current customers is attached as a supporting document (SD1). Both the business intelligence team and RAIDR utilise data feeds from secondary and community care, mental health services, urgent and primary care, prescribing and other HSCIC published datasets such as QOF, RTT etc. These data sets are provided either by the DSCRO service, downloaded directly the NHS England data catalogue or directly from provider organisations. Data delivered via the DSCRO is pseudonymised in to the CSU where the BI service and RAIDR are hosted. Published data is downloaded in aggregate form from HSCIC and NHS England websites. There will be no direct linkage between HES data records and other data already used by the CSU and in its BI tool (RAIDR). HES data may be presented alongside other data but not linked to it – for example a report may contain HES data alongside workforce statistics, weather reports etc. Typical uses for the business intelligence service and tool include • Provision of contract, performance and quality monitoring of commissioned services – this ensures CCGs are empowered with intelligence on the services they are accountable for and can undertake their statutory duties • Fully embrace clinical commissioning – CCGs have taken steps to delegate their some of their commissioning responsibility to their member practices. The RAIDR tool is used to present practice level activity and performance information allowing GP practices to assess how their local initiatives affect wider service utilization. For example, does opening their surgery later in the evening reduce the burden on A&E? Can they evidence a change in A&E usage from the point they opened later? • The national drive to progress the Better Care Fund and Vanguard alternate care models both require in-depth analytics. Access to timely information showing the impact of service transformations is key to evidencing the success of these national NHS programmes. • To support the on-going budgetary pressures the NHS is faced the business intelligence service and RAIDR offer significant support to commissioners on their QIPP programmes. Identifying service areas where the commissioner is an outlier that may then require re-procurement of a clinical service, comparisons with peer groups and best practice to understand how a change in approach might deliver a financial saving NECS is requesting HES data that will enable users of their business intelligence service and the RAIDR tool to compare themselves on a national footprint. For benchmarking, national data is needed to allow comparisons due to the number of customer organisations and number of types of organisations. This would be a major improvement on the current position where the service is limited to the data flows from the local DSCRO and published data that often is not granular enough to meet user requirements. The HES data will be utilized within RAIDR to provide a range of benchmarking dashboards and reports as required to address customer specific priorities. This may include mortality, end of life, procedures of limited clinical value, new to review ratios, readmissions etc. The ability to present a national and peer-group picture of locally defined indicators is the ambition. HES data will be presented independently of existing data flows within a bespoke dashboard as well as to supplement current reports/dashboards. For example using HES to calculate a national readmission rate to be presented on a locally fed readmission report. As well as within the RAIDR tool HES data will be used by the BI team for bespoke analytics and reporting. This will include for individual commissioners who have requested a deep-dive for a particular area and want to understand how they compare to other areas. It will also help support whole provider and health economy analysis where service re-configurations are being proposed. [Note: this is not possible with the commissioner specific slices of data provisioned via the DSCRO]
Considerable benchmarking work across the North East to look for QIPP savings, and the ability to drill down into the datasets provided by HES to understand variation at low levels of granularity. However, even if the CSU managed to bring every CCG down to the level of the best in the North East they would still be outliers nationally, as the North East is a national outlier. To this end, HES allows the CSU to place the North East into context with other sub-regions from across the country, using identical queries, to help to explain some of the limitations of QIPP schemes, but also some of the true opportunities. Using HES NECS have undertaken national comparisons against peer groups to the levels that Right Care do not do, have undertaken further deep dive reporting into the key areas of overspend, have provided comparative analysis in their routine reporting against peer CCGs, and generally make use of best national information resource. It has facilitated their CCGs to make contact with “best in class” CCGs, identified through the interrogation of HES, who have, for example, the lowest activity rates in some of the areas we’ve been looking at. This has resulted in useful discussions with those CCGs about their activity levels. The CSU's primary use of HES this year has been to benchmark their CCGs against their peer groups as identified in the Right Care packs - https://www.england.nhs.uk/rightcare/products/nhs-rightcare-intelligence-tools-and-support/, thus giving CCGs a different perspective on potential savings against peers who they would not have been able to compare against without access to HES. The benefits of this to their CCGs is improved benchmarking to help determine where they need to focus their resource for schemes within QIPP / Financial recovery, improved monitoring, to ensure schemes put in place are actually viable, and this work has helped NECS towards becoming Kite marked by Right Care. Deep-dives have been completed for individual commissioners who want to understand how they compare to other areas to help support whole provider and health economy analysis where service reconfigurations are being proposed. NECS have utilised HES data to inform baseline capacity and demand positions for the acute sector in STPs in their footprint. In particular, one of their STPs which is well progressed, is utilising HES data to inform a capacity planning model to forecast future demand and produce scenarios for future hospital configuration. This STP includes a CCG that is outwith their traditional CCG footprint, therefore the HES data provides a consistent, baseline from which this modelling can be carried out. Working with clinicians leading the STP acute sector surgical workstream to identify the most effective configuration of acute surgery across the NTWD geographical footprint. NECS are carrying out a national project to test the effect of a range of factors on patient’s decision to attend A&E departments. In order to derive the independent variable for this model, HES was used to calculate A&E Attendance rates for each of the 5000+ GP Practices within the study.
Utilising HES data for reporting will provide more accurate peer groups for benchmarking purposes rather than simply comparing neighboring commissioners as happens now using DSCRO supplied data. Benchmarking is currently restricted to data held by the local DSCRO so CCGs cannot be compared against their identifier peers as these are often elsewhere in the country. For example NECS recently delivered a project for NHS England to help improve the treatment of patients with dementia. National figures suggested the dementia registers in GP practices had patients missing. A dementia dashboard was developed within RAIDR to highlight to GP practices and commissioners where practices potentially had patients missing from their register based on secondary care SUS data – this was successful with the number of patients on dementia registers increasing however it could only be undertaken locally as NECS did not hold a national dataset. Another example is where the CSU (on behalf of a local CCG) has undertaken reporting of emergency admission rates with a view to altering patient pathways. Emergency admission rates for local providers have been used to identify best practice and pathways altered to reduce admissions. However this exercise was limited to local hospitals as NECS did not have national data available. The CCG were then able to compare their local pathways with others where readmission rates were lower with a view to changing how services were configured and commissioned locally. Ideally this would have allowed the CCG to compare their readmission rates with all commissioners/providers nationally but this was not possible without access to the full HES dataset. When the CSU come to extend/renew their agreement, evidence will be supplied for benefits achieved through the provision of dashboards/analysis to each type of customer organisation.
As described in the objectives the outputs will be two-fold: 1) Additional dashboards and reports available via the RAIDR tool to registered users covering a range of benchmarking and comparative analysis. This will include a focus on: a. Mortality b. Readmissions c. New to review ratios for outpatients d. Procedures of limited clinical value e. Falls f. Frequent flyers Allowing commissioners to compare the impact of their programmes and work streams against peer groups and nationally will help determine their effectiveness and inform future commissioning decisions. Outputs will include additions to graphs/charts showing a national and peer group figure and also tables detailing how each commissioner compares to others. The customer estimates these new dashboard will take approximately 2 months to develop and will be refreshed monthly as data is made available. 2) In addition to self-service dashboards (see above) the applicant will utilize the HES datasets to undertake various analyses both locally and in support of a range of national projects. Having the full catchment/provider data (commissioners generally only have their registered population) will facilitate accurate modelling of services and a view of complete patient pathways. Current projects where HES data would add significant value to the CSU’s services include: a. Supporting CCG vanguard applications: NECS is providing support to a number of vanguards, validating their activity/financial models and plans. Not having direct access to a standardized national dataset limits the support that can be provided. b. Service and pathway transformation: redesigning care pathways on behalf of CCGs requires access to activity data covering the entire provider with HES the only source for this. Commissioning plans must be based on accurate and complete information. The business intelligence teams within NECS would use the HES data to produce deep-dive reports and analysis on specific projects whilst ensuring small number suppression is followed for all outputs and no data is shared outside the organisation.
1. Data will be received and stored by the data management service within NECS. This is a dedicated team responsible for the organisations data warehouses and incoming/outgoing flows of data. The HES datasets will initially land in the teams secure file share before being uploaded in to an SQL Server data warehouse. Both file share and SQL server data are securely hosted within a commercial grade data centre. 2. The data management service will create derived fields based on the data received such as Ambulatory care sensitive condition flag, procedure of limited clinical value flag etc. 3. Data will be used to populate secure data cubes for use by analysts within the CSU. The data being made available will be record level but no identifiable data will be included. Only the minimum required data fields will be used to populate each cube. 4. Data will be used by the RAIDR support team to populate the relevant dashboards and reports within the RAIDR system. No patient level data will be available to RAIDR users. Small number suppression rules will be adhered to. 5. Record level HES data will not be directly linked to any other dataset. Staff follow strict rules on accessing, analysing and processing data. Only aggregate data will leave the CSU. All small numbers will be suppressed before any data is made visible to customers outside of the organisation. Small numbers will be suppressed in line with the HES analysis guide. Pseudonymised, rather than anonymised, data is required to enable calculation of benchmarked metrics on a per patient basis e.g. average number of A&E attendances per patient. For clarity, this request is for non-identifiable, pseudonymised data to flow into the data management team of North of England Commissioning Support Unit.