NHS Digital Data Release Register - reformatted
Project 1 — DARS-NIC-368543-C3J4B
Opt outs honoured: No - data flow is not identifiable (Does not include the flow of confidential data)
Sensitive: Sensitive, and Non Sensitive
When: 2016/04 (or before) — 2020/07.
Legal basis: Section 42(4) of the Statistics and Registration Service Act (2007) as amended by section 287 of the Health and Social Care Act (2012), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)
Categories: Anonymised - ICO code compliant
- Summary Hospital-level Mortality Indicator (SHMI) data split by trust and diagnosis group
- Office for National Statistics Mortality Data
- Civil Registration - Deaths
- Civil Registration (Deaths) - Secondary Care Cut
- Summary Hospital-level Mortality Indicator
The Data Recipient agrees to process the Data only for the following purposes agreed with the HSCIC: Objective for processing: To produce/analyse statistics using deaths data solely to help the NHS perform its duties. CHKS will process the SHMI data, associated HES APC data, ONS mortality data for the services listed below: • Benchmarking service for NHS Providers (Trusts) and Commissioners (CSUs and CCGs) – data are processed into a comparative database used to provide indicator level benchmarks both online and in offline reports. • Mortality profiling service for NHS Trusts – data processed and accessible at record level in pseudonymised form by individual client site only.
CHKS works directly with executive and operational teams within the NHS nationally to identify and understand variation. The use of the SHMI data has aided the understanding of the mortality data and the differences between different care systems through the analysis of in and out of hospital deaths. For example, in an organisation with multiple sites over different geographies, the data has enabled the organisation to understand and plan services to better reflect the differing needs of their population.
Expected measurable benefits to health and/or social care including target date: CHKS is currently contracted to provide services to 92 NHS organisations within England, Scotland, Wales, and Northern Ireland with contracts extending into 2018 with the primary benefit to improve patient care within the NHS. CHKS contract renewals rates within the last year are approximately 90%. The service provided by CHKS provides assurance for Trust boards and demonstrates NHS organisational commitment to continuous improvement. The services support internal analysis of performance, provide evidence for targeting improvement, demonstrate trends over time and progress made in priority areas, compare Trust performance against local targets and national peers, and engage users across client organisations. The primary benefit of using the SHMI data and the HES-ONS linked mortality data is that it enables NHS organisations to identify areas where reductions in mortality may be made and to ultimately improve the quality of patient care. CHKS anticipates being able to provide case study examples of how SHMI data are used, to support future applications. The HES-ONS linked mortality data will be used to allow analysis of both in, and out of, hospital mortality. This data will allow NHS organisations to identify particular case types where patients are dying after discharge from hospital, and identify areas where changes can be made to reduce mortality.
Specific outputs expected, including target date: Data will only be used in processed form in solely the following outputs: • CHKS live – this is a secure online portal which is accessible by authorised and authenticated users at CHKS client Trusts/CSUs/CCGs and authorised and authenticated CHKS staff. Users access the data through a range of indicator dashboards and scorecards presented at aggregate level. The online benchmarking and mortality profiling services are all accessible through the portal. Each client organisation is only given access to the specific services for which they have contracted; • Bespoke reporting – electronic or hard copy reports provided to NHS Trusts, CCGs or CSUs, providing analysis and commentary on trends in healthcare. All small numbers are suppressed in reports. It is anticipated that the above outputs will be available for use by contracted organisations within approximately 2 weeks of data being received by CHKS. Using the ONS mortality data CHKS will be able to generate summary indicators which take into account out of hospital mortality. It is anticipated that these indicators will be available within approximately 2 months of data being received by CHKS. The Mortality profiling service will display Trusts HES data for their own specific activity. Trusts can drill down to record-level for their activity for the purposes of audits and to allow NHS Trusts to review mortality cases and monitor and improve patient care (the fields that will be available in this drill down are – Admission Date, Discharge Date, Date of Death, Method of Admission, Method of Discharge, Age, Sex, Risk Prediction, Primary Diagnosis, Secondary Diagnosis). The only patient-level data field which is potentially identifiable is Date of Death (when linked with the other available fields this may make the patient record identifiable, however the Trust will only have access to data pertaining to their own patients). None of the data is linked to any client-submitted data (CHKS will not combine SHMI data supplied under this agreement with any other data that CHKS holds, other than data in the public domain) but provides information on diagnosis codes to allow meaningful audit of key conditions. No form of patient ID (HES ID) or consultant ID is included in the record-level data available to NHS Trusts. No individuals, doctors, consultants, or patients are ever identified in CHKS products, systems, or reporting using data provided by the HSCIC. Data are held in the above outputs only in pseudonymised or anonymised form. Record level raw data are never made available to any third party other than where stated in this application (where NHS Trusts can access their own activity in the Mortality Profiler)Whilst CHKS is part of the Capita Group, data are only used by CHKS for the purposes above and not shared with other organisations within the Capita Group. CHKS displays a HES data statement as per HSCIC requirements which says: “HES data re-used with the permission of The Health and Social Care Information Centre. All rights reserved.”
SHMI indicator data provided are processed using proprietary data processing software which analyses, cleanses, groups, and outputs the data into service-based patient-level databases. Any data are held only in pseudonymised form. The ONS mortality data will be linked to HES data to allow analysis of out-of-hospital deaths. Data will never be directly linked with other datasets which could allow re-identification of individuals. The CHKS live secure online system which is one of the outputs is held on CHKS servers. The servers are stored in Rackspace and Six Degrees Group datacentres which are located in England. Processed data is loaded to these servers by CHKS. Neither Rackspace nor Six Degrees Group have access to any of the outputs or data and are not involved in processing data. To minimize the amount of data held CHKS uses a rolling five years (plus year to date) data to produce the outputs required. This is to allow sufficient historic comparison of past performance. As such CHKS would only be looking to retain data back to FY 2009/10 and would delete any data held from before this period and return a certificate of destruction.
Project 2 — DARS-NIC-10891-M2Y6Z
Opt outs honoured: No - data flow is not identifiable (Does not include the flow of confidential data)
Sensitive: Non Sensitive, and Sensitive
When: 2016/04 (or before) — 2020/07.
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
- Emergency Care Data Set (ECDS)
CHKS Limited aims to produce/analyse statistics using HES data to help the NHS perform its duties. Data provided are only used by CHKS for the purposes, activities, and outputs defined in this agreement. CHKS Limited uses HES data to support and indirectly improve the provision of patient care by healthcare organisations and supporting NHS functions in England, Scotland, Wales, and Northern Ireland. NHS organisations using CHKS services benchmark and compare themselves against both national and local peers dependant on the casemix and provision of activity therefore a national dataset is required to allow such benchmarks to take place. Typically an NHS organisation will select a range of comparative providers from the national dataset, however some NHS organisations also wish to benchmark against a national acute non-specialist provider peer. In addition CHKS services allow NHS organisations to interpret and analyse national indicators, such as HSMR and SHMI, which are available at a national level. CHKS has been providing similar services to NHS organisations for over 25 years. CHKS Limited’s use of the HES data is restricted to the following: 1. Benchmarking of services for NHS providers (NHS Trusts, Local Health Boards) and NHS commissioners (CCGs,CSUs, and Local Health Boards) where data are used for creation of indicators and peer groups and are made available through an online tool and in reports; 2. Market share analysis services for healthcare providers (NHS Trusts) and commissioners (CCGs, and CSUs); 3. Data analysis toolkit services for healthcare providers (NHS Trusts, Local Health Boards) and commissioners (CCGs, and CSUs); 4. Mortality profiling service for NHS Trusts to review mortality; 5. Consultant appraisal services for NHS Trusts; 6. CHKS national Top Hospital awards celebrating success for NHS Trusts; 7. ‘Learning from the Best’ case studies for NHS organisations; 8. British Association of Day Surgery (BADS) Directory of Procedures for NHS providers.
In CHKS previous application a number of examples were set out where their client organisations were using HES data as part of the CHKS benchmarking tools to identify unwarranted variation and to target improvement work accordingly. As part of this application they have revisited these examples and highlighted just some of the improvements that have been realised in the 2017 calendar year. CHKS used an example of a secondary care provider in the South West who uses CHKS benchmarking tools to generate comparative metrics for mortality, and in particular for some of the specialist services that it operates. Using CHKS comparative analysis to direct and target improvements this organisation has seen its risk adjusted mortality outcomes improve by 7%. This equates to approximately 217 fewer deaths than would have been expected given the reported casemix of the organisation. In a second example CHKS referenced a secondary care provider in the Midlands who uses a national HES peer to review performance against a suite of indicators. Bringing clinicians into the review process of these indicators has seen substantial improvements made in a number of areas, but two of particular note is a 27% improvement in delayed discharges (using the HRG trimpoint as the discharge threshold) and an 8% improvement in overall length of stay. A large Trust in the Midlands continues to receive quarterly reporting packs derived from the CHKS benchmarking tools for key indicators such as mortality, readmissions, length of stay and other quality indicators. Using national comparison that are facilitated by the use of HES data the Trust have targeted improvements that have resulted in a 6% improvement in emergency readmissions within 28 days and also a 10% improvement in risk adjusted mortality outcomes. CHKS have made significant progress in our commissioner-based benchmarking toolkits and have recently secured a re-sign of a large commissioning support unit in London. This gives the CCGs working with this CSU continued access to the population standardised indicators that are built using HES data to provide the national coverage that is needed to generate suitable normative outcomes. The population standardised indicators include total admissions, total outpatient attendances, total A&E attendances, admitted bed days, readmissions and more. One of CCGs working with the CSU and having access to the CHKS iCommissioner toolkit have reported a 5% reduction in emergency readmissions across their commissioning population; also a 5% reduction in admissions for lower respiratory tract infections in children (for those infections that don’t generally require hospitalisation) and in addition a 12% reduction in admissions for asthma, diabetes and epilepsy in patients under 19 years old. This is a great example of using HES data to encourage collaboration between community and secondary care providers to reduce admissions and hospitalisation for patients living with chronic conditions. Using HES data CHKS assisted the British Association of Day Surgery (BADS) to update their directory of procedures that are suitable for daycase and short stay surgery. Using an associated module to monitor performance within the BADS directory a number of clients have reported progress in treating more patients as a daycase. In just one example a Trust in London treated an additional 1500 patients as daycases (across the directory of procedures) when comparing calendar year 2016 to calendar year 2017. Many clients use HES data and in particular its applicability to generate national benchmarks to assist in the monitoring of the quality of coded data. Working with a client in the South East we have used HES data to identify significant variations in the quality of coding between the two main sites within the organisation. This variation was having a particular impact on risk adjusted mortality outcomes such as the Summary Hospital-Level Mortality Index (SHMI). Targeting issues such as unnecessary transfer of patients between consultants and improving diagnosis capture in the first two episodes of the spell have seen improvements in SHMI over the last 12 months. Looking at a data quality indicator of particular interest to the Trust was the percentage of signs and symptoms being coded as a primary diagnosis in episode 2 of the inpatient spell. For the hospital site exhibiting dramatic variation to the national average, and to the other sites in the organisation, an improvement of 30% has been observed in the last 12 months.
CHKS is currently contracted to provide the above described to 92 NHS organisations within England, Scotland, Wales, and Northern Ireland with contracts extending into 2018 with the primary purpose to improve patient care within the NHS. CHKS contract renewals rates within the last year are approximately 90%. The service provided by CHKS provides assurance for trust boards and demonstrates NHS organisational commitment to continuous improvement. The services support internal analysis of performance, provides evidence for targeting improvement, demonstrate trends over time and progress made in priority areas, compare trust performance against local targets and national peers, and engage users across client organisations. NHS organisations are using CHKS services to: • Improve the quality of care; • Increase efficiency; • Increase productivity; • Monitor and reduce mortality; • Improve patient safety; • Reduce length of stay; • Reduce costs by analysing admissions; Reduce readmissions; • Improve data quality; • Monitor, analyse, and understand commissioning; • Understand service users, populations, and providers; • Plan services; • Manage risks; • Improve utilisation; • Respond to regulatory requirements. Realisation of these benefits is ongoing however to support the usage of HSCIC supplied data CHKS has made available case studies. These case studies include: • Royal Surrey County NHS Foundation Trust, who have used CHKS benchmarking tools and achieved improvements to patient safety. This was managed through the creation of indicators and benchmarks against length of stay, complications, misadventures, and mortality. Improvements to data quality were also realised. • Mid Cheshire Hospitals NHS Foundation Trust, who have used CHKS benchmarking tools and risk adjusted mortality models to identify areas where mortality indices were high and then take steps to improve the quality of care and reduce mortality. • North East London CSU, who have used CHKS benchmarking tools and national HES data to achieve improvements in provider productivity by using benchmarked data to set targets for acute Trust providers. Full case studies and more information can be found on the CHKS website at http://www.chks.co.uk/Knowledge-Base. CHKS Limited’s request for clear Consultant Code (consult) data item, for use in NHS consultant appraisal, will add further benefits as follows. Medical appraisal has been a requirement for consultants since 2001. Medical appraisal is used to support the delivery of a safe, committed, compassionate, and caring service to patients, help supervise and support doctors, and support the process of medical revalidation (Source: NHS England Medical Appraisal Policy). The addition of clear consultant code will allow CHKS Limited to provide better information to support consultant appraisal where consultants work for more than one NHS Trust. Currently those consultants who work across more than one trust are unable at present to see aggregated data in one report unless both trusts happen to be a client of CHKS. In addition many consultants now move throughout their consultant career and may often wish to have access to multiple site data which CHKS Limited have in HES but needs the Consultant Code to identify the consultant to provide trended aggregated information on performance case mix or workload. Allowing this will mean that NHS Trusts can see performance for new consultants at their first appraisal rather than relying on limited information from a few months’ work and so improving the appraisal process. Improvements to consultant appraisal will ultimately allow NHS Trusts to ensure their consultants are delivering good quality care to patients and ensure that consultants are up to date and fit to practise. The directory produced by the British Association of Day Surgery (BADS) aims to promote Day Surgery by reducing inpatient stays, and improving outcomes. The supplement adds to the information available to providers in showing how performance has changed and improved in day surgery but also shows that there still exist wide variation between providers which both providers and commissioners can use to review and optimise local performance.
CHKS produce the following specific outputs: HES data will only be used in processed form in solely the following outputs: 1. CHKS live – this is a secure online portal which is accessible by authorised and authenticated users at contracted CHKS client sites and authorised and authenticated CHKS staff. Users access the data through a range of indicator dashboards and scorecards presented at aggregate level. The benchmarking, market share analysis, data analysis toolkit, mortality profiling services, and consultant appraisal services are all accessible through the portal. Each client organisation is only given access to the specific services for which they have contracted. All users accessing CHKS live are informed to require and comply with the HES Analysis Guide; 2. Consultant appraisal reporting – electronic or hard copy reports provided to NHS Trusts providing analysis of consultant performance for appraisal. HES data used are summarised and non-identifiable and used in peer data only. Consultant benchmarks are reported independently and are not linked to individual sites. The service uses the pseudonymised consultant identifier to aggregates of Finish Consultant Episodes data, in order to show relative workload and performance indicators for consultants in peer hospitals. This is reported at anonymised and aggregated level with no patient level drill down. No other detail of consultant activity is reported. 3. Bespoke reporting – electronic or hard copy reports provided to NHS Trusts, or recognised NHS functions, providing analysis and commentary on trends in healthcare. The data will not be released outside the NHS. All small numbers are suppressed in reports in accordance with the HES Analysis Guide. 4. National awards – Trust-level aggregated indicators based on quality, improvement and best practice, and are used to determine top performing NHS organisations. Awards are held on annual basis in May. 5. Case studies – electronic or hard copy reports provided to NHS organisations. Data are provided at aggregate level only and all small numbers are suppressed. 6. BADS Directory of Procedures – National Dataset to publish alongside the guide/directory produced and published by BADS which includes the target for procedures agreed by BADS. The National dataset supplement includes data that reflects outcomes for England, with planned management intent for day surgery, and is divided into cohorts showing the percentage of procedures successfully carried out on a day case basis. Included for each procedure are aggregated indicators reporting on the performance of the top 5%, 25% and 50% of hospitals with each operation. All data is aggregated to national level and published with all small numbers suppressed. Additional Information on the above outputs • CHKS live services containing HES data are used to provide indicator and peer level comparisons in aggregated form • Within the benchmarking service NHS providers can access pseudonymised and non-sensitive record-level data for their own activity to allow providers to review benchmarks at a granular level, however all peer comparisons are at aggregated and summarised level. NHS commissioners can only view aggregated and summarised indicator level benchmarks and cannot drill down to record-level data. The Mortality profiling service allows NHS providers to access HES data for their specific activity where data is available at record level for the purposes of audits and review to allow NHS trusts to review mortality case and monitor and improve patient care. This data are not patient identifiable and is not linked to any client submitted data but provides information on diagnosis codes to allow meaningful audit of key conditions. • The Data Analysis Toolkit only allows NHS providers and NHS commissioners to see HES data aggregated in peer based reports. Users within the Data Analysis Toolkit create a tabulation by selecting from a range of available fields – the source data is at record level and the Data Analysis Toolkit then aggregates the data based on the fields the user selects. The user is then presented with the aggregated report and they do not see the record level data used to generate the tabulation. Peer based reports do not include Patient ID or Consultant ID fields. Users can download peer based reports. All users of DAT are required to accept a condition requiring adherence to the HES Analysis Guide before being permitted to run or download a Peer based report. • NHS organisations accessing these services do not have access to the HES Local Patient Identifier or the HES Consultant Identifiers through CHKS services. • Electronic or hard copy reports are provided to NHS Trusts providing analysis of consultant performance for appraisal. Consultant Code will be used in Consultant Appraisal reporting to allow consultant appraisal reports to contain activity carried out by the consultant at other NHS Trusts. This is currently not possible using pseudonymised consultant code. • The appraisal reports are made available directly to the named consultant in each trust or to the appraisal manager/Coordinator/revalidation responsible officer or medical director in the Trust where the consultant’s main contract is held. consultant’s work can be seen in other trusts but in summarised and aggregated form and not at patient level – the consultant report summarises activity, length of stay, day cases rates, complications, readmissions, and mortality indicators. Consultant reports will not be made available to the public by CHKS and will solely be provided to NHS Trusts that are clients of CHKS. • The clear consultant code field will only be used for the Consultant Appraisal objective, processing and outputs. Other relevant supporting information No individuals, doctors, consultants, or patients are ever identified in CHKS products, systems, or reporting using data provided by the HSCIC. HES data are held in the above outputs only in pseudonymised form and are never associated with other datasets held in CHKS systems. Record level data are never made available to any third party organisation unless specifically stated. Whilst CHKS Limited is part of the Capita Group only aggregated data are only used by CHKS Limited for the purposes above and not shared with other organisations within the Capita Group. CHKS displays a HES data statement wherever HES data are used. The statement says: “HES data re-used with the permission of The Health and Social Care Information Centre. All rights reserved.” This statement is present on all CHKS live pages, any extracts downloaded from CHKS live, and all bespoke consultancy reports. The CHKS live secure online system is held on CHKS Limited servers. The servers are physically stored in Rackspace and Six Degrees Group datacentres which are located in England. Processed record-level HES data is loaded to these servers by CHKS. Neither Rackspace nor Six Degrees Group have access to any of the outputs or data; these organisations provide physical locations to host the servers and network infrastructure but the servers are exclusively managed and used by CHKS Limited.
HES data provided are processed using proprietary data processing software which analyses, cleanses, groups, and outputs the data into service-based patient level databases. Any HES data are held only in pseudonymised form and are never directly linked with other datasets which could allow re-identification of HES data. As well as HES, CHKS Limted process other datasets (directly submitted patient data, and publicly available datasets – PLACE, SHMI, RTT,Friends & Family Test, PROMS, Reference Costs, Staff Survey, Patient Survey, Cancer Waits, CDIFF/MRSA, Safety Thermometer, CQC Intelligence Monitoring). These other datasets are not directly linked to HES but are available as indicators. A user could view an indicator derived from HES data (e.g. Average LoS, Mortality) on the same screen as indicators derived from the other datasets mentioned. Using the patient level databases held CHKS Limited generates different sets of benchmarks, indicators, risk models, peer groups, and records which are then used for the services defined below: 1. Benchmarking service to NHS providers and NHS commissioners – HES data are processed into an aggregated comparative database used to provide indicator level benchmarks both online and in reports. 2. Market share analysis services – HES data are processed into an aggregated provider or commissioner based comparative market share databases. 3. Data Analysis toolkit service – HES data are processed into a pseudonymised patient database on which NHS providers and NHS commissioners can run summary peer reports. 4. Mortality profiling service – HES data processed and accessible at record level in pseudonymised form for client site only. 5. Consultant appraisal service – HES data processed and used to create aggregated consultant peer groups for comparative analysis. This is the only activity that utilises the Consultant Code field. 6. CHKS national Top Hospital awards – HES data are processed and used to create aggregated indicators held at Trust level. 7. NHS case studies – uses the aggregated comparative HES database also used for the benchmarking services; 8. BADS Directory of Procedures supplement - uses the aggregated comparative HES database also used for the benchmarking services. CHKS uses five years worth of HES data to produce the outputs required. This is to allow sufficient historic comparison of previous years performance. Once the 2014/15 annual refresh data has been received and processed CHKS will delete the 2008/2009 HES data.