NHS Digital Data Release Register - reformatted

Northumbria Healthcare NHS Foundation Trust

Project 1 — DARS-NIC-249035-R2Z5Y

Opt outs honoured: Yes - patient objections upheld (Section 251 NHS Act 2006)

Sensitive: Non Sensitive, and Sensitive

When: 2020/11 — 2020/11.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(7)

Categories: Identifiable

Datasets:

  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Critical Care
  • HES:Civil Registration (Deaths) bridge
  • Civil Registration - Deaths
  • Hospital Episode Statistics Admitted Patient Care

Objectives:

The NIVO Study, (Non-Invasive Ventilation Outcomes Study), is a 10-centre prospective trial led by Northumbria Healthcare NHS Foundation Trust, assessing outcomes in hospital and after discharge in patients who are ventilated for exacerbations of COPD (chronic obstructive pulmonary disease). Northumbria Healthcare NHS Foundation Trust are the sponsor for the NIVO study and sole data controller under this Agreement. The other 9 centres are not involved in decisions regarding the collection or processing of the data under this Agreement, further details can be found below. COPD is a common lung disease and accounts for a large number of hospital admissions each year in England. COPD is frequently complicated by episodes of acute worsening of respiratory symptoms, termed ‘exacerbations’. Severe exacerbations of COPD may lead to respiratory failure. Non-invasive ventilation (NIV) is a method of supporting the patients’ normal breathing efforts, using a mask fitted to their face connected to a ventilator, and can be lifesaving during exacerbations of COPD complicated by respiratory failure. Compared to invasive ventilation, which involves placing a tube in the patient’s main airway, sedation is avoided, complications such as infection are less common, outcomes better and a much larger proportion of patients are considered eligible. The earlier NIV is started, the better the outcomes for patients. Historically NIV was provided in intensive care units, but it has been shown to be safe and effective when delivered in non-intensive care settings. Therefore, over recent years there has been an appropriate expansion in NIV use and the decision to initiate NIV is now undertaken by a variety of clinicians, including non-specialists. Despite widespread availability, there is significant under-use of NIV. Clinicians initiating NIV need to be able to quickly identify which patients would benefit from NIV and which patients may require alternative therapies. This complex decision requires an assessment of an individual’s chances of survival, should NIV be provided. Of concern, even specialist clinicians are unduly pessimistic about outcomes following NIV, which may lead to patients being inappropriately denied treatment. The research team at Northumbria Healthcare NHS Foundation Trust are currently deriving a predictive tool, using indices that are readily available and designed to be easy to apply at the bedside, to predict inpatient mortality in exacerbations of COPD requiring assisted ventilation. In a pilot study, researchers have identified novel predictors of in-patient death and are confident that an effective tool can be created. Before a predictive tool can be used in clinical practice, it is vital to ensure it is generalisable to other hospitals and other geographical locations. The primary aim of the data requested under this Agreement, is to validate this tool. The benefits of this work are numerous; primarily researchers anticipate increasing the use of lifesaving treatment (NIV) by improving clinical decision making. Identifying predictors of 6 month mortality is valuable for discharge planning and can improve patient-centred decision-making regarding future care, including provision of palliative care. As researchers are creating a clinical tool to predict death, it is imperative for NIVO to include those who are sickest (and therefore at highest risk of death). Approaching these patients to consent into a trial would be impossible in all situations, often they are confused, drowsy and unable to consent. Not including this patient group, however, would unduly bias the study and results from the study would be inaccurate in this patient group. Therefore support under section 251 of the NHS Act 2006 has been fully approved to allow the Northumbria Healthcare NHS Foundation Trust to access data for this cohort of patients without their consent. Patient and public opinion has been sought with strong support voiced for the project. Northumbria Healthcare NHS Foundation Trust began tracking COPD patients in October 2016 after their initial treatment with ventilation and then following patients for one year following discharge from hospital, to assess recovery or deterioration and death. The final patient's 1 year follow up ended in February 2019. A cohort of 844 patients are tracked by the NIVO study. These patients have been identified by clinicians across several sites in the UK. At each site, consecutive, unique patients meeting the selection criteria have been identified by prospective screening of the admission unit and areas where NIV is provided by clinicians. Survival data is collected at each site up to 12 months post discharge and this data is sent to the research team at Northumbria Healthcare NHS Foundation Trust. The data collected includes routine data about health resource utilisation: namely hospital readmission and mortality data. This information is easy to get when the patients re-present to a NIVO hospital trust, but when patients present to other trusts, or when they die in the community, the data controller is unable to access data for the patient. This problem has become apparent as the study has progressed. In order to accurately assess and draw conclusions on health resource utilisation, and on co-morbidity burden in this group, complete data is important. HES and mortality data is collected routinely on these patients and held by NHS Digital. It is comprehensive; including the number of hospital attendances with diagnostic code, as well as number, duration of and cause for admission (including to critical care). Mortality data will allow the data controller to assess the definitive cause of death in patients who have been treated with NIV for COPD, thus facilitating the ability to draw accurate conclusions about disease progression and prognosis. Access to the data outlined above will help inform future management, especially treatment of co-morbid conditions, in this patient population. It may result in increased recognition of co-morbid conditions and therefore improved referrals and access to specialist care; increased provision of respiratory specialist care in the community and additionally support post discharge. The potential benefits of analysis of this data is in reducing morbidity and mortality in this group of patients as a whole, with significant improvement in health status for COPD patients across the country. It will also improve communication with patients about how their illness may affect them over the months after their discharge. For information about the wider study, a longitudinal part of the NIVO study will be undertaken which is not relevant to the data requested under this Agreement. For the longitudinal component of the trial, the data controller will seek consent from patients to follow-up their data. The purpose of this longitudinal study is to help inform a better understanding of outcomes, should a patient survive to discharge as this will help inform discussions about future care planning. Researchers intend to follow up patients with particular focus upon quality of life after assisted ventilation. This longitudinal study is not relevant to the data requested under this Agreement. The data controller has approval from the local ethics committee for the NIVO study (16/NE/0213) and considers that risk to the public by dissemination of this data is highly unlikely. Patient information has been publicly displayed within the recruitment centres as advised by CAG, and public involvement in the Project Management Group has received positive feedback. Only results which will be aggregated with small numbers suppressed in line with the HES Analysis Guide will be released into the public domain. The data processing is in line with Article 6(1)(e) ‘processing is necessary for the performance of a task carried out in the public interest’. COPD affects up to 3 million people in the UK alone and with the progressive shift in the age of the population, these numbers are likely to increase over the next decade. As deemed by the Confidential Advisory Group, there is strong public interest in this activity, public interest is in line with Article 9(2)(j) ‘processing is necessary for scientific or historical research purposes’. The research group believes the moral principle of guiding treatment with objective outcome data is paramount. Prior to commencement of the trial, a patient group was convened and proposed methods were discussed and the trial protocol was amended and then endorsed by the patient group. Similarly, prior to the application for data from NHS Digital, patients in the NIVO were contacted and support was gained. Additionally, throughout the active recruitment period of the study, the study group held trial steering committees with an lay representative present (this was a patient who has COPD). Ethical approval was sought from the regional NHS ethics committee, and was granted to collect data (including 1 year outcomes such as readmission and cause of death) without individual patient consent. The rationale behind not requiring individual consent was to ensure that the data controller could include the sickest patients (those most likely to die, and very unlikely to be able to give informed consent), as these patients are of the utmost importance when trying to model predictions of mortality. These ethical issues (no individual patient consent) were endorsed by the local ethics committee but also in the initial, and the NHS Digital, patient groups. This approach has been taken by several other groups developing prognostic research tools, further justifying its acceptability. The data under this Agreement will be used in the results of the NIVO study to describe accurate health resource utilisation, co-morbidity and mortality (including cause of death) in the year following admission requiring ventilation. The data controller requests one export of data covering the year after the NIVO patients index admission (of which the admission date may have been at any time from Oct 2016 to Feb 2018) and discharge date from Oct 2016- May 2018. Only Northumbria Healthcare NHS Foundation Trust will process the data under this Agreement. Other organisations involved in recruitment to the NIVO study are: - Gateshead Health NHS Foundation Trust - South Tynside NHS Foundation Trust - Leeds University Hospitals NHS Trust - Nottingham University Hospitals NHS Trust - Oxford University Hospitals NHS Trust - Guys and St Thomas' NHS Foundation Trust - Taunton and Somerset NHS Trust - Plymouth University Hospitals NHS Foundation Trust The Newcastle University is involved in the wider study in two ways: 1) A statistician from Newcastle University has reviewed the statistical plan for the study in an advisory capacity. The statistician does/will not have access to any of the data under this Agreement nor make any decisions regarding outputs. 2) The Chief Investigator holds professorship at Institute of Cellular Medicine, Newcastle University but is a substantive employee of Northumbria Healthcare NHS Foundation Trust. The CI takes responsibility and control of the data requested under this Agreement in their role as Consultant in Respiratory Medicine at Northumbria Healthcare NHS Foundation Trust. Newcastle University is not a data controller under this Agreement as no personnel acting under the auspices of this organisation are involved in the decisions regarding the data requested under this Agreement. Analysis of the data will be undertaken by a research doctor working within the CI's research group. All research staff are substantive employees of the data controller and all staff are NIHR GCP trained. The NIVO project is funded in part by physician initiated competitive grants from Philips Respironics, the ResMed Foundation, Pfizer OpenAir and Bright Northumbria. All funders provided competitive, charitable grants for clinician-initiated research into COPD. Funders have not, nor will have, any input into the design, development, conduct of, analysis of, or intention to publish results of the study. Funder 1. Philips Respironics awarded a competitive, charitable grants for clinician-initiated research. Philips Respironics produce medical devices including non-invasive ventilators. Whilst indirect, if the conclusion of this research recommends a change in the increased provision of ventilation this could have an indirect benefit (likewise, if the study recommends the opposite, it may be of financial detriment). Funder 2. Pfizer OpenAir provides charitable grants for research into COPD. These grants are awarded independently of Pfizer Inc, who produce pharmaceutical products. There is no clear potential financial gain to be made from Pfizer or Pfizer OpenAir from this research. Funder 3. The ResMed Foundation provides charitable grants for clinician-initiated research into ventilation. These grants are awarded independently of ResMed Corporation. ResMed Corporation produce medical devices including non-invasive ventilators. Whilst indirect, if the conclusion of this research recommends a change in the increased provision of ventilation this could have an indirect benefit (likewise, if the study recommends the opposite, it may be of financial detriment). Funder 4. Bright Northumbria is a hospital-based charity at the Northumbria Healthcare NHS Foundation Trust. Its board awards grants for initiatives perceived to be of benefit to patients. There is no possible financial benefit to Bright Northumbria from this research.’ These funders have had no input into the design, conduct or running of the study. They will have no input into the analysis of data, results or decisions to present or publish, nor will they have access to any data other than results which will be aggregated with small numbers suppressed.

Expected Benefits:

This data will allow the data controller to analyse re-admission cases, frequencies and causes of death in a sick patient cohort, in a way not done before. This data, when linked to the initial inpatient data as well as quality of life data collected, may allow researchers to recognise groups of patients, by their clinical phenotype, who do well or badly after NIVO treatment. Recognition of these groups would allow more objective deployment of healthcare staff/resource to improve the quality of life of patients. For example, if a large proportion of patients are admitted with heart failure in the year after discharge it may be appropriate to start actively screening ventilated patients for this in the outpatient department, resulting in earlier diagnosis and optimisation. The outputs could also be of interest to commissioners, general practitioners and policy writers. Without these outputs, there would not be any comprehensive knowledge of all admissions/deaths in the COPD population and therefore the results would be less reliable. Additionally, and equally importantly, the outputs may allow clinicians to have better informed discussions with patients about their prognosis and disease severity. This will facilitate things like more appropriate advanced care planning. The NIVO study is currently the subject of two PhD's through the Institute of Cellular medicine, Newcastle University. The first of these doctoral theses (to which this Agreement is not relevant) is planned for completion in 2020. All the analysis for the second PhD ((which will include data gained from this agreement) is being done through Northumbria Healthcare NHS Foundation Trust on the NIVO database. Only aggregate data will be used for any analysis and subsequent write up of the thesis (with only aggregated data used in any reports/presentations/thesis). The thesis will be submitted to Newcastle University at the end of the project (akin to writing a research paper and submitting to a journal), but Newcastle University do not have access to any data under this Agreement. The second PhD should be completed by autumn 2020. Following the analysis of this data, preparation of manuscripts for publication will be performed and the data controller would hope to have the first ready by Autumn 2020.

Outputs:

Results from the NIVO study including reports, submission to high ranking peer-reviewed journals, abstracts and presentations at conferences are expected in the two years following completion of the study and analysis of results (main results to be published by a target of 2021). This is the largest trial of the COPD population to have occurred in the UK and has potential to influence national (and international) practice with the use of the predictive modelling tool and data on year long outcomes. Interim results from the study (not including NHS Digital data) have already been presented at national and international conferences (The British Thoracic Society Winter Meeting in 2017, 2018 and 2019; and the European Respiratory Society International Congress 2018 and 2019) and generated significant interest. Reports will also be presented to commissioners and providers, general practitioners and NHS policy makers. Participants and the general public will be made aware of the results through the Northumbria Healthcare NHS Foundation Trust website. All outputs will only contain results in highly aggregated format and as statistical summaries and measures of association. Small numbers will be suppressed in line with the HES Analysis Guide. Record level information will not be released to any third party.

Processing:

Data to allow identification in HES and mortality data sets will be transferred from the data controller to NHS Digital. This will be limited to: - study number - gender - post code - NHS number - date of birth - date of discharge from index hospital admission (to allow calculation of 1 year post discharge data) The study number, along with pseudonymised HES and mortality data, will be returned to the data controller from NHS Digital. Further to this, the data from NHS Digital will be added into the NIVO database. Only pseudonymised data will be added, and all of the participant identifiers will be secured separately, in a limited access folder and password protected files only available to the chief investigator and sub investigators at the Northumbria Healthcare NHS Foundation Trust. No identifying data will be used or referenced in the analysis of results, only aggregated data (with small numbers suppressed) will be used in the write up, presentation or dissemination of results from the trial. Recruiting NIVO sites will not be involved in any communication with NHS Digital. All episodes of HES data are requested; this is because following admissions with COPD requiring ventilation, patients may be admitted for other conditions which are attributable in part to COPD (such as heart failure, cor pulmonale, anxiety etc) which reflect their poor functional status and overall burden of illness. Increasingly COPD is recognised within a pattern of multi-morbidity, something which has been recently highlighted in the national COPD audit program. If researchers accurately capture all re-admissions, not just those COPD related, researchers may recognise patterns which allow the team to proactively target/screen and optimise other medical conditions (such as heart failure, osteoporotic fractures etc), and would improve the accuracy of communication about prognosis/outcome with patients. Data flow: - Each recruiting NIVO site will send patient identifying data to the data controller (via encrypted NHS.net email). - The data controller will collate the information and allocate a NIVO study number for each participant. - The data controller will send identifying participant data to NHS Digital - No further flow of identifying data will be required - NHS Digital link HES and mortality data to participant data and return pseudonymised data along with the study number. No identifying details are returned to the data controller. - Data controller links pseudonymised NHS Digital data with pseudonymised participant data, linked only by study number (all identifying details will be held separately as detailed above) - Data controller performs analysis of merged data and subsequent drafts of manuscripts written - Results and manuscripts disseminated to individual sites for approval/comment - Publication/presentation of results of study at conferences and in medical literature. Re-identification of data post linkage to study number would only be accessible to the data controller as they are the only site with the collated data set. The NIVO database (in which statistical analysis occurs) does not and will not have any identifiable data included, and once the results of readmission/death/resource utilisation is merged there should be no requirement for re-identification or re-linking with identifying data, nor will the data controller attempt to re-identify any participant. Data collation, linking and analysis of results will only be performed by the Chief Investigator and his research group who are all substantive employees of Northumbria Healthcare NHS Foundation Trust. In all outputs no identifying data will be included, only aggregated results. Aggregate post code analysis of disease prevalence by ward may be presented as an output, however the number of patients (~3million in the UK) and frequency of admission to hospital in the general COPD population would render identification of individuals impossible from this data. All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract i.e.: employees, agents and contractors of the Data Recipient who may have access to that data).