NHS Digital Data Release Register - reformatted

Royal College Of Anaesthetists projects

303 data files in total were disseminated unsafely (information about files used safely is missing for TRE/"system access" projects).

🚩 Royal College Of Anaesthetists was sent multiple files from the same dataset, in the same month, both with optouts respected and with optouts ignored. Royal College Of Anaesthetists may not have compared the two files, but the identifiers are consistent between datasets, and outside of a good TRE NHS Digital can not know what recipients actually do.

MR1386 - National Emergency Laparotomy Audit — DARS-NIC-355855-R4G6G

Type of data: information not disclosed for TRE projects

Opt outs honoured: Yes - patient objections upheld, N, Identifiable, Anonymised - ICO Code Compliant, Yes, No (Section 251, Section 251 NHS Act 2006)

Legal basis: Section 251 approval is in place for the flow of identifiable data, 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(7), Health and Social Care Act 2012 – s261(7), Health and Social Care Act 2012 – s261(7); National Health Service Act 2006 - s251 - 'Control of patient information'., Health and Social Care Act 2012 - s261(5)(d)

Purposes: No (Academic)

Sensitive: Non Sensitive, and Sensitive, and Non-Sensitive

When:DSA runs 2019-01-01 — 2021-12-31 2017.06 — 2024.03.

Access method: One-Off, Ongoing

Data-controller type: HEALTHCARE QUALITY IMPROVEMENT PARTNERSHIP (HQIP), HEALTHCARE QUALITY IMPROVEMENT PARTNERSHIP (HQIP), NHS ENGLAND (QUARRY HOUSE)

Sublicensing allowed: No

Datasets:

  1. Hospital Episode Statistics Admitted Patient Care
  2. MRIS - Cohort Event Notification Report
  3. MRIS - Cause of Death Report
  4. MRIS - Flagging Current Status Report
  5. Hospital Episode Statistics Critical Care
  6. Civil Registration - Deaths
  7. Demographics
  8. HES-ID to MPS-ID HES Admitted Patient Care
  9. Hospital Episode Statistics Admitted Patient Care (HES APC)
  10. Hospital Episode Statistics Critical Care (HES Critical Care)
  11. Civil Registrations of Death

Objectives:

Emergency abdominal surgery (or emergency laparotomy) is associated with significant morbidity and mortality worldwide. The aim of The National Emergency Laparotomy Audit (NELA) is to enable the improvement of the quality of care of patients undergoing emergency laparotomy by providing high quality comparative information of the clinical practice and outcomes of all NHS providers of emergency laparotomy in England and Wales. NELA is a national clinical audit commissioned by the Healthcare Quality Improvement Partnership (HQIP) as part of the National Clinical and Patient Outcomes Programme (NCAPOP).

HQIP have commissioned the Royal College of Anaesthetists (RCoA) to deliver the audit. RCoA are working in partnership with the Clinical Effectiveness Unit of the Royal College of Surgeons (RCS) and therefore make up the NELA project team. HQIP act as data controllers for the national clinical audit but do not have access to any data collected or analysed by the RCoA and RCS who are the data processors. The analysis by the NELA project team will only involve pseudonymised (with exception to full Date of Death and Cause of Death) datasets that combine information submitted by NHS hospitals and data supplied by NHS Digital / ONS. Members of the project team work at either the RCoA or Clinical Effectiveness Unit at the RCS, however analysis of the NHS Digital, ONS, and an extract of pseudonymised NELA data (including full Date of Death obtained via hospitals) will only be processed and stored at RCS.

The RCoA and RCS wish to link the patient records submitted to NELA with the Hospital Episode Statistics (HES) records for those patients. The NELA records relate only to an individual admission, and by linking to inpatient HES data, the Audit will be able to provide more precise and relevant information to NHS hospitals by allowing the RCoA and RCS to describe longer term outcomes (e.g., readmission rates) and to improve RCoA’s and RCS’s risk-adjustment models by using the extensive information on comorbid conditions held within HES. (eg to calculate the Charlson Comorbidity score)

The RCoA and RCS also wish to link the patient records submitted to NELA with the Office for National Statistics (ONS) Death Register on a quarterly basis to enable the Audit to monitor changes in postoperative outcomes (both short and longer-term mortality) for those patients. Access to this linked information will support this national clinical audit to improve the quality of care within NHS hospitals for a high-risk patient group.

The objectives of the Audit are:
1. To enable secondary care providers to improve the delivery of care to patients undergoing emergency laparotomy using information produced by the audit;
2. To provide comparative information on the organisation of care by providers of Emergency Laparotomy.
3. To provide comparative information on patient outcomes following surgery for Emergency Laparotomy.
4. To facilitate the development of effective change (quality improvement) initiatives and thereby spread examples of best practice and help local providers make the best possible use of audit results


In summary, the purpose of this request is to support national clinical audit, quality improvement within hospital, and research on methods to monitor surgical outcomes.
The NELA team is planning to request patient level clinical data from Intensive Care National Audit and Research Centre (ICNARC).

Yielded Benefits:

Benefits derived from the research so far: - Data is being used by CQC on their Dashboard (no record level data - all aggregated with small numbers suppressed) - Data being used on MyNHS website (no record level data - all aggregated with small numbers suppressed) - Work has started on introducing a BPT for Emergency Laparotomy using the audit data. - Headlines from reports of the last 3 years: o national 30-day mortality rate falling from 11.8 per cent to 10.6 per cent over three years, representing approximately 300 patients lives saved each year compared to 2013-2014. o Making recommendations in the reports for local participating sites to follow o Running regional NELA Quality Improvement workshops to assist hospital sites in using their local data to improve patient care Reports/Publications Created: - 3 National Patient Audit Reports published in 2015, 2016, 2017 - 2 Organisational Audit reports published in 2014 and 2017 - Quarterly hospital reports provided to participants since the start of 2017 Conferences/Presentations Attended: There are multiple as the NELA Project Team attend many meetings throughout the year. Here are a few of the more recent ones: - -Emergency Laparotomy Collaborative Event - 2017 - -HQIP - Patient First Event - 2017 - -RCoA Patient Safety Conference - 2017 - -AAGBI Winter Scientific Meeting – January 2017 - - Sweden emergency surgery meeting - 14th March 2017 - - Royal Society of Medicine - Anaesthesia & Surgery Mtg – April 2017 - - Age Anaesthesia Association Meeting 25th May 2017 - - ASGBI Surgical Conference – May 2017 - - HSRUK Symposium 2017 - July 6th-7th, 2017

Expected Benefits:

The NELA audit is highly relevant to current clinical practice and publications will allow widespread disseminate of the findings amongst health professionals. Linkage to the HES/ONS data allows the Audit to report more extensively on patterns of care beyond the initial hospital admission and longer-term outcomes, such as 90-day mortality. The Audit is able to examine issues such as readmission rates and the most common reasons for these post-discharge complications, e.g.: respiratory complications and anastomotic leaks.

The audit will produce useful indicators that describe the standard of care in a variety of clinical areas. The indicators will identify NHS providers that are performing well and those requiring improvement to the quality of care received by patients.
Ongoing improvement in the processes of care and clinical outcomes should lead to a reduction in the postoperative mortality rates and thus an overall improvement in patient outcomes. Outcomes will be measured by re-auditing individual sites and therefore regular data linkage would be required. It is hoped that this improvement in care would be identified by the end of the currently proposed commissioned audit period (Dec 2018).

The intended audience for the audit annual reports are clinicians, healthcare professionals, Medical Directors, Chief Executives, audit managers, commissioners, NHS England, public and patients. Trusts will use the process indicators and outcomes reported in the annual reports to assess their care against national standards and benchmark against other NHS trusts. This will enable providers to identify areas requiring improvement and take action which in turn will provide a benefit to patient care.

Reporting will identify whether NHS trusts are meeting national guidance such as NICE recommendations and will identify variations in the provision of care. The Audit outcomes such as postoperative mortality are risk-adjusted and any potential outlying trusts are identified as part of the Audit outlier policy.

Any Trust showing as an alarm will be notified which will allow for investigation into the cause; this can be attributable to either data quality issues or clinical practice. This notification will enable the trust to address the cause and either review the data submitted to the Audit or their clinical practice. Any resultant improvements in clinical practice will directly impact on improvements in patient care.

The trust profiles are publically available, providing transparency and enabling patient choice.

Publishing in peer-reviewed journals will allow greater discussion of the strengths and weaknesses of the results, and will provide the benefit of peer-review of the work from third parties.

It is anticipated that the reports produced as a result of the audit will contribute to clinical guidance and national policy.

Outputs:

The linked dataset will be a product of this process and will enhance the quality of the comparative data for the audit in subsequent years.

The NELA is commissioned to produce a "State of the Nation" annual report each year. The first 2 reports utilising patient level information were published in June 2015 & June 2016 and are available to view on the NELA website. Subsequent reports are scheduled to be published yearly.

In order to more widely disseminate the findings of the audit, additional scientific publications will be produced. These outputs will be in the form of peer-review articles and conference presentations.

The results of the audit will also be disseminated at professional medical conferences and in peer-reviewed journals (e.g. BMJ, BJA, ASGBI journal, AAGBI journal) at the time of the launch of the report or shortly after. Publications related to the Audit methods (e.g., a risk-adjustment model) rather than information of clinical practice and outcomes will be published on an ad hoc basis.

In response to participant feedback a Quality Improvement Report Dashboard has been created on the NELA Online Web Tool to assist sites with audit data collection and to promote local Quality Improvement work (Local hospital data can only be viewed by registered local hospital participants). The Quality Improvement Report Dashboard will only provide local units with aggregated information to compare their performance against a national average. The figures available to each unit will be based on their own local data (supplied by the units) and the Dashboard may therefore present small numbers on some occasions. Each individual user has their own login to the webtool which gives them access to only their own hospital local data. To access the webtool they require a username and password.

The first phase of the Dashboard which has now been launched focuses on Case Ascertainment and Patient Demographics. The Patient Demographics section allows local participants to view some basic information on their hospital's population of patients undergoing emergency laparotomy, and how it compares to the audit-wide average. It focuses on characteristics such as patient age and operative urgency.

The Case Ascertainment aims at increasing case completeness and submission by providing a monthly list of cases entered/completed/not completed.

The next phase of the Dashboard will focus much more on Quality Improvement, feeding back key QI indicators and comparing hospital's local data with national audit-wide averages. Some of the measures reported back will include; Documentation of risk, Direct Admission to critical care etc.

The Royal College of Anaesthetists are currently in the process of developing the QI dashboard and hope to make further additions in the next few months.

All outputs will be aggregated with small numbers suppressed and will follow the ONS/HES guide on reporting.

Processing:

The RCoA are the principal data processors for NELA and manage the extraction of the records from the NELA IT system.

RCoA will send the file of patient identifiers and the NELA ID to NHS Digital for linkage to HES and ONS fields. Pseudonymised files (of which includes full date of death and Cause of Death (text)) from NHS Digital will contain the HES and ONS fields with the NELA ID variable added. The pseudonymised files of HES / ONS data (including Date of Death and Cause of Death) will be received by the RCS and held on their secure data server.

In all cases, the data received from NHS Digital will not be linked back to the identifiable NELA database held at RCoA. An extract of pseudonymised data of which includes date of death is taken from the NELA database and sent to RCS. This data will be linked to the HES-ONS data via the NELA ID. Date of Death from the NELA database is provided as this includes an important data quality step. There are potential missed linkages if RCS and RCoA do not have this information when processing the data at RCS. This also helps to validate the data entered into NELA. No data provided by NHS Digital is sent to RCoA to correct fields in the NELA database.

A copy of the de-identified data fields along with the unique NELA ID will be analysed by project team members from both RCS and RCoA at the RCS Clinical Effectiveness Unit only. The only identifiable fields received by RCS and RCoA analysts are ONS Date of Death and Cause of death. The full Date of Death is required to be able to calculate survival at multiple time points (30 day, 90 day, etc.). There will be no capacity for RCS / RCoA analysts to use the ONS Date of Death to identify any individual patients.

Analysts from RCS / RCoA who work on the de-identified data set do not have access to the identifiable data set held within the NELA IT system and managed by the RCoA, nor the list of patient identifiers sent to NHS digital for linkage purposes. All individuals with access to the de-identified data are substantively employed by RCoA or RCS. The list of patient identifiers sent to NHS digital is only accessible by a senior member of the RCoA. RCoA or RCS will not be linking HES/ONS data with any other dataset (apart from an extract of NELA). Linkage with any other datasets would be subject to a future application and would be supported by an appropriate legal basis.

The majority of the analysis involving the de-identified linked patient dataset will be conducted by the RCS statisticians who form part of the NELA Project Team. The remainder would consist of statisticians from RCoA would be involved in some of the analysis of the patient-level dataset and will be located at the RCS Clinical Effectiveness Unit to undertake this work. In either case, all individuals with access to the data are substantively employed by either RCoA, or RCS and are required to sign a Confidentiality Agreement before access is granted.

Perioperative Quality Improvement Programme (PQIP) — DARS-NIC-63347-R8J2M

Type of data: information not disclosed for TRE projects

Opt outs honoured: No - data flow is not identifiable, Anonymised - ICO Code Compliant, Identifiable, No (Reasonable Expectation, Consent (Reasonable Expectation))

Legal basis: Informed Patient consent to permit the receipt, processing and release of data by the HSCIC, Health and Social Care Act 2012 – s261(2)(c), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(2)(c), , Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(2)(b)(ii)

Purposes: No (Academic)

Sensitive: Non Sensitive, and Sensitive, and Non-Sensitive

When:DSA runs 2018-11-03 — 2021-11-11 2017.12 — 2022.01.

Access method: One-Off, Ongoing

Data-controller type: ROYAL COLLEGE OF ANAESTHETISTS, ROYAL COLLEGE OF ANAESTHETISTS, UNIVERSITY COLLEGE LONDON (UCL)

Sublicensing allowed: No

Datasets:

  1. Hospital Episode Statistics Admitted Patient Care
  2. Hospital Episode Statistics Critical Care
  3. Office for National Statistics Mortality Data
  4. Civil Registration - Deaths
  5. Civil Registration (Deaths) - Secondary Care Cut
  6. HES:Civil Registration (Deaths) bridge
  7. HES-ID to MPS-ID HES Admitted Patient Care
  8. Civil Registrations of Death - Secondary Care Cut
  9. Hospital Episode Statistics Admitted Patient Care (HES APC)
  10. Hospital Episode Statistics Critical Care (HES Critical Care)

Objectives:

Perioperative Quality Improvement Programme (PQIP) is a new study run by the National Institute of Academic Anaesthesia, Health Services Research Centre (NIAA-HSRC), and the Royal College of Anaesthestists (RCoA), and the Health Foundation, with a planned initial duration of 5 years.

NIAA and NIAA Health Services Research Centre are departments within RCoA. University College London is the sponsor and the Health Foundation is funding the study. Neither the University College London nor the Health Foundation is acting as Data Controllers in relation to this application. University College London Hospitals NHS Foundation Trust is a joint data processor with RCoA. All processing of data is carried out at RCoA.

PQIP will measure risk-adjusted morbidity and mortality, as well as process and patient-reported outcome data in patients undergoing major surgery. The dataset has been informed by previous systematic and structured reviews, and over 60 UK NHS hospitals have so far volunteered to take part.

RCoA will collect objective risk, process and outcome data on patients during their inpatient stay. The PQIP database will only collect data for the primary admission when patients undergo their planned operation. By linking to Hospital Episode Statistics (HES) data and Office of National Statistics (ONS) mortality data at record level RCoA will be able to analyse a more complete dataset. This is necessary to track adverse outcomes which occur after discharge from hospital (e.g. readmission within 30/90 days of surgery - from HES data; longer term mortality - from ONS).

The objectives for PQIP are:
Primary Objective -
To measure the rate of postoperative complications after major inpatient surgery in the UK and how it varies between hospitals. RCoA primary analysis will measure risk-adjusted variation between providers (comparing observed: expected ratios) in morbidity, mortality and failure to rescue rates.

Secondary Objectives –
(a) What is the failure to rescue rate in NHS hospitals and how does it vary?
(b) What is the relationship between short-term complications and longer-term health related quality of life (HRQOL),
and can longer-term HRQOL be improved through reducing postoperative complications?
(c) Can the quality of care for patients undergoing surgery in NHS hospitals be improved through the feedback of data to clinicians and managers, leading to improvements in complications and failure to rescue?

Aggregated outcome data will be reported back to participating hospitals to inform of their outcomes (hospital length of stay data, mortality data), together with data collected within the PQIP dataset.

In summary, the purpose of this request is to support a UK wide clinical study, involving quality improvement within hospitals, and research on methods to monitor surgical outcomes. Through measuring process and outcome measures for patients undergoing major surgery the aim is to improve the quality of care for patients throughout the UK.

The NIAA-HSRC is a body which sits within the remit of the Royal College of Anaesthetists. UCL (University College London) are acting as sponsors for the study, with the joint research office (a partnership between UCL and UCLH). The Health Foundation have supplied grant funding for the project but have no involvement in the design, day to day running of the project, or the outputs created from the study.

Yielded Benefits:

The dissemination of data to the Perioperative Quality Improvement Programme (PQIP) supports the project in meeting its primary objective ‘to measure and report risk-adjusted complication, patient reported outcomes and mortality rates after major surgery’. The project provides hospital level data on a quarterly basis to NHS hospital trusts to help them identify areas where they may be able improve the quality of care for patients undergoing major surgery. The dissemination has supported PQIP in reporting risk-adjusted outcomes after surgery across the UK. Outcome reporting will allow comparison of the quality of care across NHS trusts and support improvement in the quality of care. The main aim of PQIP is to improve risk adjusted outcome from major surgery across the UK. Around 1.5 million patients undergoing major surgery every year in the NHS. Complications may occur in up to 25% of them. By reducing complications and improving outcomes PQIP will improve the quality and efficiency of major surgery in the NHS. The dissemination also supports PQIP in comparing the sample of patients recruited to those undergoing eligible procedures in the wider NHS. By doing this PQIP can validate its strategy and demonstrate that the outcomes reported are representative. Outputs from PQIP have helped increase the knowledge and understanding of outcomes after surgery across the UK. It has also helped to identify where significant variation occurs between outcomes, and through measurement of hospital structures and processes will offer some insight and explanation as to why it occurs. Hospital levels reports support local quality improvement with rapid feedback of data. Annual reports provide an overview and national picture. NHS Digital data has supported the RCoA to produce national risk-adjusted mortality outcomes. As data is being processed to fulfill a "legitimate interest "Article 6(1)(f); the following has been demonstrated: • Purpose – to improve care for patients undergoing major surgery across the NHS. Reduce complications, deaths, and improve efficiency within the NHS. • Necessity – NHS Digital data is needed in order to the aims of PQIP. The strain of direct data collection would risk the project failing to meet its aims and would not be feasible. • Balancing – the potential benefits of PQIP to the population of the UK are significant. Over 1.5 million major operations take place annually. PQIP is recruiting in over 100 hospitals. It is not possible to collect the level of data present in the critical care and hospital episode statistics database using local data collectors. Follow-up mortality outcomes would also not be possible due to the strain on resources. Without this information PQIP would not be able to meet its stated aims and objectives. The NHS Digital data dissemination has supported PQIP in meeting these aims. Future Potential benefits include: • Local quality improvement – improvement in processes and outcomes at a local level. Produced by regular quarterly feedback of hospital level data to NHS trusts taking part in PQIP. Two annual collaborative events were held in 2018 with further planned for 2019. • Supporting the development of local quality improvement collaboratives. • Dissemination – o Hospital levels reports (produced every 3 months) o Annual report – produced April 2018 (https://pqip.org.uk/FilesUploaded/PQIP%20Annual%20Report%202017-18.pdf) o Publications in medical literature 1.5 million patients undergo major surgery each year. By reducing complication and hospital length of stay there are potential benefits to the whole of the NHS. Cost savings by reducing length of stay by 1 day for only 100,000 patients (<10% of those eligible) would amount to £40 million per year. The benefit will to individual NHS trusts and to the NHS as a whole. The RCoA measure hospital processes and outcomes as part of the study. Hospital length of stay is directly collected, together with complication and mortality outcomes and all stated benefits will be achieved within the timescale of the PQIP study (planned to run until December 2021). Outputs have supported the increase in NHS trusts taking part in PQIP. The study now has over 100 hospitals taking part, having recruited over 15,000 patients to date. The annual report (https://pqip.org.uk/FilesUploaded/PQIP%20Annual%20Report%202017-18.pdf) gives and overview of national performance and quarterly hospital level reports provide rapid feedback on both process and outcomes to trusts and supports them in local QI work.

Expected Benefits:

PQIP will produce indicators that describe the standard of care for patients undergoing a variety of major surgical procedures throughout the UK. These indicators will identify NHS providers that are performing well and those needing to improve the quality of care provided to patients.

Ongoing improvement through feedback of this data should lead to a reduction in the postoperative morbidity and mortality rates and thus an overall improvement in patient outcomes. Outcomes will be measured by ongoing data collection at hospital sites throughout the project duration and therefore regular 6-monthly data linkage would be required. It is hoped that the data collected will lead to steady, continuous improvement in care throughout the whole duration of PQIP, but within this period, a clear signal of improvement by the publication of the second annual report planned for March 2019 .

Linkage to the HES/ONS data allows PQIP to report more extensively on patterns of care for patients undergoing major surgery beyond the initial hospital admission to longer-term outcomes, such as 90-day mortality. PQIP will be able to examine issues such as complication rates across different types of surgery and hospital trusts in order to identify areas for improvement in patient care.

The intended audience for the PQIP annual reports are clinicians, healthcare professionals, Medical Directors, Chief Executives, audit managers, commissioners, NHS England, public and patients. Trusts will use the process indicators and outcomes reported in the annual reports to assess their care against national standards and benchmark against other NHS trusts. This will enable providers to identify areas requiring improvement and take action which in turn will provide a benefit to patient care.

PQIP has been designed as a quality improvement programme (based on work previously done in the US - NSQIP). RCoA will be reporting performance over time back to NHS trusts through quarterly reports. As part of the study there will also be qualitative research work, which will assess the uptake of PQIP, and look at barriers to adoption. RCoA will also use feedback from trusts on the reporting structure and metrics and adapt these if required to maximise impact. RCoA have initially deliberately chosen not to benchmark trusts against each other, instead choosing to look for continuous improvement from trusts. We will provide national references for metrics as that data becomes available.

Reporting will identify whether NHS trusts are meeting national guidance such as NICE recommendations and will identify variations in the provision of care. Trusts showing poor performance will be notified to allow for investigation into the cause; this can be attributable to either data quality issues or clinical practice.

RCoA will report metrics back to trusts, and look for continuous improvement. As part of the qualitative research taking place within PQIP the study will assess quality improvement interventions undertaken by trusts, and offer local support for initiatives to improve quality.

The annual reports and trust profiles will be openly available, providing transparency and enabling patient choice.

The initial planned quarterly reports will be made available to trusts taking part in PQIP. The format of larger annual reports is yet to be decided, but these are likely to be openly available. There is also a desire to make the quarterly reports openly available but RCoA are assessing the views of local teams.

Publishing in peer-reviewed journals will allow greater discussion of the strengths and weaknesses of the results, and will provide the benefit of peer-review of the work from third parties. It is anticipated that the reports produced as a result of the audit will contribute to clinical guidance and national policy.

Outputs:

The purpose of the NHS Digital linkage is to reduce the data collection burden of local sites with regards to readmission rates, allow RCoA to collect mortality data, and to validate their sampling strategy.

The NHS Digital linked data will be used to produce quarterly/annual reports and academic outputs.

Quarterly reports will be produced reporting type of surgery or hospital level outcomes. These will be targeted to specific groups involved within the PQIP research team (nurses, research team, anaesthetists, surgical team, and hospital managers). A full PQIP report and executive summary will be produced annually, with the first planned for March 2018.

In addition, research manuscripts using PQIP data will be submitted to open access peer reviewed journals. Wider dissemination to the surgical and anaesthetic profession will be achieved using the resources of the RCoA and other stakeholder Royal Colleges and the NIAA-HSRC, including websites, press releases, written and electronic communications. Different resources will be used to disseminate information to different stakeholders, using a multi-media approach and lay representation to ensure effective communication to the public.

RCoA will be submitting journals to Anaesthesia, British Journal of Anaesthesia, The New England Journal of Medicine, or The Lancet. There may also be publications to other journals that are produced using the PQIP data. The target dates for these publications will be in years 2-3 of the study onwards (November 2018-2019 onwards).

Processing:

The RCoA are the main data processors for PQIP and manage the extraction of the records from the PQIP IT system (which is stored on UKFast servers).

The RCoA are requesting two separate linkages the first will include sending patient identifiers for the consented cohort to NHS Digital for linkage and the second request involves a linkage via OPCS procedure codes where no patient identifiers will be sent to NHS Digital and RCoA will receive back pseudonymised HES data. No patient identifiers will flow back to NHS Digital for this second linkage.

For the first data linkage the RCoA will send the file of patient identifiers and the PQIP ID to NHS Digital for linkage to HES and ONS fields.

Patients who RCoA supply identifiable data to NHS Digital for the linkage (those included in the PQIP study) have been consented. The second linkage (via OPCS procedure codes) will not have consented, this linkage does not involve identifiable data being shared or linked. No ONS mortality data is being requested for this second linkage.

In order to allow this, a file (P) containing these patient identifiers only will be extracted from the full dataset hosted in the PQIP webtool, and will be sent securely to a trusted Data Linkage Service (TDLS). For mortality tracking data and HES data, this would be NHS Digital. File (P) will contain the following identifiers:
• PQIP study identifier
• NHS number
• Date of Birth
• Sex
• Postcode

For each patient in the file, NHS Digital will identify the matching ONS or HES ID. NHS Digital will then return to the Royal College of Anaesthetists a ‘look-up’ file (L) containing only the PQIP identifier and the HES or ONS ID identifiers, and a MATCH_RANK field which indicates the strength of the match. An extract of pseudonymised HES or ONS mortality data will then be requested from NHS Digital for all the list of IDs contained in file (L).

The file (L) will be placed in the secure RCoA server accessible only to the project data manager. It will then be used to link the pseudonymised ONS or HES data to the pseudonymised PQIP data for analysis. The pseudonymised PQIP extract will not contain NHS number, postcode or date of birth. Patients will be labelled with the PQIP identifier only. The pseudonymised dataset is analysed to produce results for PQIP reports and other PQIP outputs. All outputs produced will be pseudonymised aggregated level data with small numbers suppressed in line with HES analysis guidelines.

RCoA will also request a second linkage as part of the same data sharing agreement, this is separate to the consented cohort. This will be to enable the PQIP team to obtain denominator data for the PQIP patient cohort. This data will be used to ensure the sampling technique for PQIP is representative of the total population of patients undergoing major surgery across the UK. This second linkage will be done via OPCS procedure codes. The PQIP team will provide NHS Digital with a list of OPCS procedure codes that are included within the PQIP study. RCoA will then ask for pseudonymised, non-sensitive record level data for those patients including HES data such as hospital length of stay, 30-day readmission and 30-day/90-day mortality flags.

The pseudonymised data NHS Digital supply will be used by the PQIP research team (the PQIP research team consists of NIAA-HSRC employees who are substantive employees of the RCoA and also substantive employees of UCLH) to produce hospital reports and academic outputs. The only access to patient identifiable data held within the PQIP database will be from substantive employees from the RCoA and only for the purpose of completing the NHS Digital linkages. The pseudonymised data (only containing PQIP identifier) will be passed on to the PQIP research team for data analysis.

RCoA will not link the HES/ONS data with any other dataset other than the PQIP data. Linkage with any other datasets would be subject to a future application and would be supported by an appropriate legal basis.

RCoA also plan to link to PEDW in Wales, and to NHS Scotland. These other linkages will be performed separately to the NHS Digital linkage and will not involve the sharing of NHS Digital data with Wales or Scotland. The linkages to central NHS held data will only involve the transfer of data for patients recruited in those nations, so for example there will be no data transferred to NHS Digital for patients recruited in an NHS Scotland institution. NHS Digital data will not be disseminated to Wales or Scotland for the purpose of this data linkage.

Once RCoA have received data from NHS Digital, this will be linked via PQIP patient study number to the data held within the PQIP webtool, and this will be exported (pseudonymised with only PQIP patient number, no other identifiable data) to the PQIP team who will be performing the data analysis.

The analysis of pseudonymised linked data will be conducted by the RCoA statisticians/fellows who form part of the PQIP research team. The PQIP data analysis/research team are either substantive employees of UCLH or the RCoA.

All processing of ONS data will be in line with ONS standard conditions.
All outputs and publications contain only aggregated data with small numbers suppressed in line with the HES Analysis Guide.

MR1477 - EPIdemiology of Critical Care After Surgery (EPICCS) — DARS-NIC-88623-F2H1Q

Type of data: information not disclosed for TRE projects

Opt outs honoured: Yes - patient objections upheld, Anonymised - ICO Code Compliant, Yes, Identifiable (Section 251 NHS Act 2006)

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(2)(b)(ii), Health and Social Care Act 2012 - s261(5)(d)

Purposes: No (Academic)

Sensitive: Sensitive, and Non Sensitive, and Non-Sensitive

When:DSA runs 2019-01-10 — 2020-01-09 2019.06 — 2019.07.

Access method: One-Off

Data-controller type: ROYAL COLLEGE OF ANAESTHETISTS, UNIVERSITY COLLEGE LONDON (UCL)

Sublicensing allowed: No

Datasets:

  1. MRIS - Flagging Current Status Report
  2. MRIS - Cause of Death Report
  3. Hospital Episode Statistics Admitted Patient Care
  4. Hospital Episode Statistics Critical Care
  5. Hospital Episode Statistics Admitted Patient Care (HES APC)
  6. Hospital Episode Statistics Critical Care (HES Critical Care)

Objectives:

The Sprint National Anaesthesia Projects (SNAPs) are 'snapshot' evaluation studies of clinical activity and patient-centred outcomes that are important and relevant to both patients and anaesthetists.

The 2nd Sprint National Anaesthesia Project: Epidemiology of Critical Care provision after Surgery (SNAP-2: EpiCCS) will describe the epidemiology of perioperative risk and outcome, and critical care referral and admission after inpatient surgery in the UK. It also aims to examine whether planned postoperative critical care admission is effective as an intervention to reduce postoperative morbidity.

After surgery, complications can sometimes occur, including mild ones like nausea, and more major ones such as infections, and very rarely death. One way which may help to prevent complications in some patients is to admit patients to a Critical Care Unit (CCU) after their operation, where they can receive more intensive nursing support, or particular treatments not available on other hospital wards. Critical care would normally only be considered for people who are having either a very big operation, or who have a number of significant background illnesses. Previous research studies have shown that the proportion of patients admitted to CCU after surgery differs between hospitals and countries. The study has been conducted to try and uncover some of the reasons for these findings, and to see if Critical Care admission after surgery improves patient recovery after surgery.

The overall aim of the study is to help improve care for people undergoing surgery.

The purpose of the proposed data linkage is to enable the RCoA to answer research questions about the associations between postoperative morbidity, postoperative admission to a critical care unit, and longer-term outcomes for patients undergoing major (inpatient) surgery.

Background:
The admission of high-risk patients to critical care after surgery is a recommended standard of care. Nevertheless, poor compliance against this recommendation has been repeatedly demonstrated in large epidemiological studies. It is unclear whether this is due to reasons of capacity, equipoise, poor quality clinical care or because hospitals are working creatively to create capacity for augmented care on normal surgical wards. Importantly, the clinical effectiveness of critical care as an ‘intervention’ is uncertain, and the threshold for admission to critical care (i.e. the risk profile of patients who might benefit) is currently based on expert opinion rather than research.

The EPIdemiology of Critical Care after Surgery (EPICCS) study aims to address these uncertainties. The study is a one-week observational cohort study including all patients who underwent inpatient (overnight stay) surgery between March 21st - 27th 2017. All patients had prospective data collection on risk factors, surgical procedure and postoperative outcomes including the primary outcome of morbidity (measured using the Postoperative Morbidity Survey on day 7 after surgery) and secondary outcomes including length of stay and inpatient mortality. The epidemiology of patient characteristics, processes and outcomes will be described once analysis is conducted. Inferential techniques (multilevel multivariable regression, propensity score matching and instrumental variable analysis) will be used to evaluate the relationship between critical care admission and postoperative outcome. The protocol for the study has been published in the British Medical Journal: (Moonesinghe SR, Wong DJN, Farmer L, Shawyer R, Myles PS, Harris SK. SNAP-2 EPICCS: the second Sprint National Anaesthesia Project EPIdemiology of Critical Care after Surgery: protocol for an international observational cohort study. BMJ Open. 2017 Sep 1;7(9):e017690. )

University College London (UCL) and University College London Hospitals NHS Foundation Trust (UCLH) Joint Research Office, is the office which helps to administer research to both partner organisations. The Royal College of Anaesthetsists (RCoA) and UCL are the data controller for this study, as joint decisions on how to use the data rest with the RCoA and the Chief Investigator of the Project - substantively employed by University College London.

EPICCS is being jointly led (in terms of determining the study and how the analysis will be conducted) by staff at UCL and the National Institute of Academic Anaesthesia’s Health Services Research Centre (NIAA-HSRC) - who for this project are based at the RCoA. The NIAA was established in March 2008. Its vision, with respect to anaesthesia and related specialties, is to improve patient care by promoting the translation of research findings into clinical practice; develop and maximise its academic profile within the healthcare profession, NHS, universities and major research bodies; facilitate high profile, influential research; facilitate and support training and continuing professional education in academia. It is based at the RCoA (who are joint data controller and processor) , and UCL/UCLH Joint Research Office but with no role in determining how the data will be used or processed - that's decided by UCL and RCoA and funded by the Royal College of Anaesthestists, the UCL Surgical Outcomes Research Centre and the National Institute for Academic Anaesthesia. The study has been funded by these organisations on the basis of scientific merit and potential to improve patient outcomes.

Aim:
1. To describe the epidemiology of critical care admission after surgery
2. To determine the clinical effectiveness of planned critical care admission after surgery

Research Questions pertinent to this application
1. Do previously validated risk stratification tools accurately predict postoperative mortality?
2. Does immediate critical care admission reduce postoperative morbidity and mortality?
3. Does short-term morbidity independently predict longer term mortality and other adverse outcomes (e.g. hospital readmission)?

Study design

EpiCCS is a one-week, prospective observational cohort study of patients and anaesthetists in NHS hospitals. It incorporates the following:

Data will be collected by perioperative anaesthetists on all patients undergoing inpatient surgery in participating UK hospitals for one week. Postoperative morbidity will be recorded for patients who remain in hospital on Day 7 after surgery. Mortality data will be collected through linkage with Hospital Episode Statistics data facilitated by NHS Digital. The dataset will also include patient risk factors, and questions about clinical decision-making and resource availability related to critical care referral and admission. The main EPICCS patient study will include all patients undergoing inpatient surgery (elective or emergency) during the study week.

Additionally, an organisational questionnaire for each hospital will be completed to describe structure and process in those institutions, the number and reasons for any cancellations of surgery during the study, and critical care unit occupancy at regular time-intervals throughout the one-week data collection period. The epidemiology of perioperative risk stratification, postoperative care and patient outcome will be described. Multivariable regression, instrumental variable and propensity score matched analyses will be conducted to ascertain the clinical effectiveness of postoperative critical care admission in reducing adverse outcomes after inpatient surgery.

A survey of anaesthetists and surgeons on their attitudes and behaviours regarding risk prediction and postoperative critical care admission. All anaesthetists and surgeons who undertake perioperative care for inpatient surgery during the study period will be invited to participate.

At the time of data-entry onto the EpiCCS webtool, confidential patient information will be retained and stored securely in their original format within the database, however different database access privileges (dependent on usernames and passwords) will apply to different users of the database:
• Local investigators within NHS Trusts will have access to their own full datasets, including patient identifiable information.
• The central EpiCCS study team will only have access to an anonymised dataset for analysis. Among the patient identifiers, only sex will be used for analysis. In this dataset the NHS number will be replaced by a unique study patient identifier; Date of Birth will be converted to Age on date of surgery, and trimmed to month and year of birth; Postcode will be converted to PCT, SHA of residence, and the Office for National Statistics Lower Super Output Area, which allows the allocation of the Index of Multiple Deprivation.
• The Royal College of Anaesthetists (Deputy Chief Executive Officer) will extract the required patient identifiable data from the study database onto a password protected Excel spreadsheet, and email this securely to NHS Digital to perform data linkage. NHS Digital will return to the Royal College of Anaesthetists a file containing the HES and MRIS data and a study id. The NHS Digital data will be linked to the EPICCS data and a pseudonymised dataset will be accessed by UCL and RCoA for analysis.


EPICCS was a one-week, prospective observational cohort study of patients undergoing inpatient surgery in participating hospitals. The study recruited 22,993 patients for one week in March 2017 (patient recruitment phase), in a total of 245 NHS Hospitals throughout the UK with a planned follow-up of 10 years (to track longer-term survival). Recruitment has now therefore finished.

Eligibility criteria
All UK NHS hospitals which undertake inpatient surgery were eligible to take part. All UK hospitals are eligible but it is only those hospitals with patients in critical care during the one week outlined in the application that participate in the study.

Inclusion criteria:
Adult (>=18 years) patients undergoing surgery or other interventions who were expected to require overnight stay in hospital which require the support of an anaesthetist. These would include all procedures taking place in an operating theatre, radiology suite, endoscopy suite or catheter laboratory for which inpatient (overnight) stay is planned, including both planned and emergency/urgent surgery of all types, Caesarean section, surgery for complications of childbirth, endoscopy and interventional radiology procedures.

Exclusion criteria:
Patients who indicated they do not want to participate in the study; Ambulatory surgery; children (<18 years); non-surgical obstetrics; ASA-PS grade VI; non-interventional diagnostic imaging (e.g. CT or MRI scanning without interventions); emergency department or critical care interventions requiring anaesthesia or sedation but no interventional procedure

The purpose of patient-level data linkage is to:
- be able to track longer term mortality via the MRIS mortality tracking system;
- to report hospital readmission (rates and reasons) through linkage with HES data.

HES APC and CC data are required to determine rates and reasons for readmission to hospital after the initial discharge following surgery. Hospital readmission is a complication of surgery which should be included in any outcome analysis.

Full date of death from patients in the cohort is required to track patient outcomes and for survival analysis modelling (which requires accurate time to death following surgery duration to perform).

The RCoA are also requesting an extract of tabulated data. This purpose of this is to obtain total number of surgical procedures carried out in England during the study period in hospitals participating in the 2nd Sprint National Anaesthesia Project: EPIdemiology of Critical Care provision after Surgery - during 21st March 2017 and 27th March 2017. The number of procedures for each procedure code should be tallied for each hospital. These data will then be compared against the total number of cases recruited to the SNAP-2: EPICCS study in order to identify how accurate the sampling strategy has been in capturing data on a representative number of surgical operations being conducted across England. This will be provided under a separate agreement issued by NHS Digital.

The PhD work will form part of the wider study, PhD students are often affiliated with large studies and their work will concentrate on a small section of the study. These are the study questions that will directly relate to the PhD thesis:
1.How do clinicians determine the risk of postoperative mortality in clinical practice?
2. Do previously validated risk stratification tools accurately predict postoperative mortality?
3. On what basis do clinicians refer patients for planned postoperative critical care?
4. What factors influence whether patients actually receive planned postoperative critical care?

Expected Benefits:

The EPICCS study will evaluate research questions related both to policy and clinical practice.

Policy: Do we have sufficient numbers of critical care beds to meet the demand from surgery and to provide a high quality service?

Are surgical patients in hospitals where there is increased demand for critical care beds disadvantaged?

Clinical practice:
Do current methods of risk prediction accurately predict short and longer term outcome after surgery?

Is critical care an effective intervention to prevent postoperative complications?

Answering these questions will provide opportunity to substantially improve quality of perioperative care and our understanding of the factors which influence patient outcomes after major surgery.

Publishing in peer-reviewed journals will allow greater discussion of the strengths and weaknesses of the results, and will provide the benefit of peer-review of the work from third parties. It is anticipated that the reports produced as a result of the study will contribute to clinical guidance and national policy. In particular the relevant stakeholder organisations involved in study planning will receive reports of our findings which will directly influence policy recommendations and guidelines issued by them which the Department of Health and NHS England look to for future workforce and infrastructure planning. The Royal College of Anaesthetists is a member of the Academy of Medical Royal Colleges, and therefore provides expert guidance to the Department of Health and other NHS bodies on strategic matters concerning perioperative medicine, surgery and anaesthesia. The Chief Investigator for this study is also the associate national clinical director for elective care at NHS England, and therefore has high-level access to key decision makers within the NHS in England. The study has benefitted from collaboration with over 2,800 anaesthetists, surgeons and research professionals in NHS Trusts throughout the UK, who are all directly invested in the study’s results. Any outputs will therefore be looked upon keenly by healthcare professionals closely involved in patient care at the coalface.

Outputs:

This study will yield several research manuscripts which will be submitted to open access peer reviewed journals. Wider dissemination to the surgical and anaesthetic profession will be achieved using the resources of the RCoA and other stakeholder Royal Colleges and the NIAA-HSRC, including websites, press releases, written and electronic communications. Different resources will be used to disseminate information to different stakeholders, using a multi-media approach and lay representation to ensure effective communication to the public. Authorship of all manuscripts will be determined by the CI, with the agreement of the Project Team (UCL and RCoA members), and comply with ICJME standards.

Although the study has not been commissioned it has received grant funding and support from the NIAA and (Association of Aneathetists of Great Britain and Ireland Project Grant), the RCoA and the UCLH/NIHR Biomedical Research Centre (UCL Surgical Outcomes Research Centre). The study has further been adopted onto the NIHR Clinical Research Portfolio. Therefore all study findings will be reported to these respective organisations. Other stakeholder organisations have been involved in study planning, including the Faculty of Intensive Care Medicine and Intensive Care Society. These organisations will also receive a report of the study findings in order to support future healthcare policy recommendations.

The results of this study are likely to yield significant patient benefit in terms of outlining patient accessibility to postoperative critical care and ways this might be improved, and outcomes of the study will be disseminated by the Royal College of Anaesthetists, which has a Lay Committee panel and Patient, Carer & Public Involvement & Engagement (PCPIE) group, both of which have significant patient and public involvement, and will advise on how best to communicate the findings in language understandable by the public. Furthermore the Royal College of Anaesthetists has a significant Communications Department which has experience in disseminating public education materials.

The study has yielded one published manuscript so far, and two more are in the final stages of preparation before submission to peer-reviewed journals. The first published manuscript (Wong, Harris, Moonesinghe for the SNAP-2 collaborative, British Journal of Anaesthesia, Sep 2018; https://www.ncbi.nlm.nih.gov/pubmed/30236235) led to substantial coverage in the medical and lay media (e.g.: https://www.bbc.co.uk/news/health-45432538 )

The initial findings and study methodology have already been presented at several national / international meetings including:
SG ANZICS (Asia Pacific Intensive Care Forum) (May 2018)
EBPOM (Evidence Based Peri-Operative Medicine) Conference (July 2018)
Intensive Care Society - State of the Art (December 2018)
Royal College of Anaesthetists – 2 conferences so far

The RCoA anticipate that analyses based on linked data will yield several more high-impact publications which have the potential to change clinical practice and/or policy.

Data will not be used for sales and marketing purposes.

All reports / results will be published on the study website and be available to participants. All published material will contain aggregated outputs only with small number suppression in line with the HES Analysis guide. No identifiable patient information will be contained within the published material.

In addition, the research will be disseminated by:
- Publications in open access peer reviewed scientific journals
- Face to face presentations
- Live-action & animated video content on an open-access YouTube channel
- Social media
- Conference presentations
- Professional literature e.g. Royal College of Anaesthetists literature

The aim will be to publish the first manuscript which uses NHS Digital data within 12 months of receipt of the data.

Processing:

Patient level data
Data from patients undergoing surgery in the UK between 21 - 27 March 2017 were collected and entered onto a secure online database (EPICCS IT system) controlled by the Royal College of Anaesthetists. The EPICCS IT system is stored on UK Fast servers. This data included patient identifiable information, however data exports used by researchers from this database are pseudonymised on export. Patient identifiable information is only accessible by the Data Controller for the study at the Royal College of Anaesthetists (Deputy Chief Executive Officer). Patient identifiers will be extracted from the database, which will contain the following identifiers in order to perform linkage with HES and MRIS (Mortality) Products from NHS Digital:

- EPICCS Study identifier
- NHS number
- Date of Birth
- Sex
- Postcode

RCoA will send the data extract to NHS Digital securely. For each patient in the data extract, NHS Digital will identify the matching HES episodes and the Mortality data from the MRIS products. NHS Digital will then return to the Royal College of Anaesthetists a file containing only the requested data and the EPICCS identifier (Study ID).

This file with NHS Digital linked data will be placed in the secure RCoA server accessible only to the project data manager, a substantive employee of RCoA, and will not contain NHS number, postcode or date of birth. The dataset will be analysed by the RCoA to produce results for EPICCS reports and other EPICCS outputs, but will not be relinked or used in conjunction with any other data to re-identify patients or to perform any other analysis outside of this study.

The data returned from NHS Digital will be linked to the EPICCS data and this pseudonymised dataset will be accessed by UCL and RCoA for analysis.

Data storage
The study will use UKFast to store data. The server is hosted by UKFast in a secure datacentre. All traffic passes through Cisco equipment including Anomaly Detection Systems (ADS), Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS). This is clustered across multiple locations. The hardware is protected by a Cisco Firewall with full access controls enabled. UKFast will carry out annual security audits. These audits will inspect the system for any vulnerabilities or threats that could allow hackers to destroy or damage the system. Each UKFast datacentre is fully powered, secure, resilient and equipped to meet the project demands. The company has a track record in delivering to many private and public sector clients including the NHS and has demonstrated a strong awareness of the need to protect systems and data from both physical and virtual threats. UKFast has attained ISO-27001:2013 certification for their Information Security Management System and ISO 9001:2008 for their Quality Management System. They are PCI compliant for all client transactions.

UCL and RCoA have overall responsibility for the research and the outcomes of the study and are joint data controllers.