BISTRO Trial application for NHS Digital data for data linkage

The "BioImpedance (BI) Spectroscopy To maintain Renal Output" - the BISTRO Trial - is a randomised controlled trial funded by the National Institute for Health Research (NIHR), the research funding arm of the NHS.

Most patients who develop kidney failure choose unit-based haemodialysis treatment. Dialysis removes waste products and excess fluid from the blood when the kidneys stop working properly. Haemodialysis involves diverting blood to a machine to be cleaned.

One of the main functions of dialysis is to control the amount of fluid in the body. Too much fluid can lead to raised blood pressure that damages the heart and increases the risk of stroke, and may cause fluid to collect in the lungs leading to breathing difficulties. Too little fluid causes dehydration, cramps and low blood pressure and more rapid or complete loss of any remaining kidney function. Bioimpedance (BI) is a simple, bedside measurement giving information about body composition, specifically how much excess fluid is present. Clinicians can use this to guide how much fluid should be removed from the body with the normal clinical assessment of the amount of fluid in the body, but it is not known if this results in better decisions and outcomes for patients.

This trial aims to test whether taking regular measurements with a bioimpedance device, which gives information about body composition, improves outcomes for people who have newly started haemodialysis treatment for kidney failure. In particular, the trial aims to see if this helps patients maintain their remaining kidney function, as this is associated with improved survival, fewer symptoms of kidney failure, fewer side effects of dialysis treatment and a better quality of life including confidence in managing their health, and cost benefit analysis.

People starting haemodialysis as an outpatient with some remaining kidney function were invited to participate in a clinical trial that compares current best practice with the same but additionally guided by regular bioimpedance measurements. The study has randomised 438 consented patients from about 34 dialysis units across the UK. This agreement refers to the 381 consented individuals from England only.

Patients that have given their informed consent were randomly allocated into two study groups. Other than allocation of the intervention under investigation (which participants have given informed consent to be allocated through randomisation), there will be no decision making based solely on automated means.

The BISTRO trial was designed to use routine data collection where possible, and for this purpose data linkage is required with UK Renal Registry and Hospital Episodes Statistics (HES), via NHS England, databases for the collection of dialysis-related information to the renal registry and events. People treated with in-centre dialysis unfortunately experience many complications, admissions to hospital (both planned and unplanned) and frequently require interventions and clinic attendances from departments external to the kidney dialysis unit. While some of these events are collected as Serious Adverse Events via paper case report forms (CRFs) for trial monitoring purposes, many are not. Detailed knowledge of these is required, in particular for the Health Economic analysis, which is fully funded by the National Institute for Health Research Health Technology Assessment (NIHR HTA) Programme.

Examples of these events include episodes related to vascular access failure and associated interventions, cardiovascular events and their associated interventions, the length and cause of hospital admissions and death.

The data linkage ensures more reliable capture of inter-current events and procedures and post-trial follow up which will allow the study team to determine if there is a legacy effect and link early dialysis events to long-term outcomes.

COLLECTION OF TRIAL DATA: Good clinical practice-trained research staff provided the eligible study participants with information about the trial in the form of a participant information sheet. If the patients agreed to participate, they will have consented by signing a consent form. During the trial duration, the research nurse completed paper forms containing:

(a) administrative data (with identifiable information),

(b) clinical data and

(c) intervention delivery related data.

The research nurse also asked the patient to complete a patient quality of life questionnaire. The data collected by the participating sites (NHS Trusts) and is stored securely for the 438 participants randomised into the trial. These documents were returned to Keele Clinical Trials Unit.

The trial administrator and trial manager Keele Clinical Trials Unit received the paper documents from the recruiting sites and entered them into two types of databases: (a) a management database and (b) a data entry database. Hard copy paperwork is held securely within an access-controlled environment within Keele University. Electronic and hard copy data is pseudonymised as soon processing is complete. Pseudonymised clinical data is linked by a BISTRO Study ID.

DATA REQUESTED: The datasets requested are as follows:

• Hospital Episode Statistics (HES) Admitted Patient Care (APC - required for the evaluation of secondary outcomes: significant events and cost-effectiveness of the intervention

• Hospital Episode Statistics Critical Care (CC) - required for the evaluation of secondary outcomes: significant events and cost-effectiveness of the intervention, particularly in light of COVID-19

• Hospital Episode Statistics Outpatients (OP) - required for the evaluation of secondary outcomes: significant events and cost-effectiveness of the intervention

• Civil Registration (Deaths) Secondary Care Cut – required for long-term legacy effects.

Keele University will send a cohort of 381 individuals identified only by the BISTRO Study ID and the identifiers NHS Number, forename, surname, and date of birth, to NHS England for linkage to HES data. All study participants have provided informed consent to allow linkage and collection of this data.

Data is required from when the trial completes follow up in 2021. The study is intended to run for 4 years, and will require outcomes data for the period from April / May 2017 to August 2021.

Study Participants were recruited from centres UK-wide and data is therefore requested across all the devolved nations. This agreement refers to data collected in English Hospitals only.

To collect these data using research nurses would have added considerably to the cost of the trial. The use of routine data makes this trial affordable and therefore possible. All study participants have been provided with information about the nature of the trial and how their personal and sensitive data will be processed, and have given their explicit consent. NHS England has considered the adequacy of the consent materials in relation to the application and has determined on balance that data flow is compatible with the consent.

DATA MINIMISATION: only data items necessary for the analysis of the trial have been requested; specifically: Primary Health Economic analysis and pre-specified secondary analyses. All the relevant trial data has been collected using Case Report Forms, except for routinely clinical data collected by external data registries.

For trial participants who have consented to data linkage, routine clinical data collected by units for the Renal Registry returns will be transferred to the CTU for incorporation into the trial database (from 2017 to end of follow-up plus a final download 5 year years post first randomisation for legacy data only) in the form of an electronic downloads (following appropriate testing procedures to ensure data integrity). This includes data collected for individual dialysis sessions (e.g. pre and post weights, blood pressure dialysis prescription), haematology and biochemistry results, and treatment modality timelines, using the Renal Registry Dataset v4.2 (UKRR website). Admission and discharge dates, diagnostic and procedural codes will be obtained from HES datasets via NHS England (or its equivalent body for devolved nations).

The data has been requested in line with inclusion and exclusion criteria as described in the study protocol, focussing on adults aged >18 years, within 3 months commencing centre-based maintenance haemodialysis as an outpatient (Day 0) due to advanced kidney disease CKD stage 5 who have given consent to participate on the trial. There were no reported pregnancies in the trial, as expected, as this is mostly an older population.

The sole data controller for this agreement is Keele University. The data will be processed at Keele Clinical Trials Unit which is part of the Keele University and by the University of Warwick for Heath Economics analyses. The Chief Investigator and all co-investigators accessing NHS England data are substantive employees of the Keele University or Warwick University.

The UK Renal Registry will be providing data to the Keele University on kidney disease outcomes and a separate data sharing agreement is in place for this.

The funder is NIHR HTA. NIHR do not determine the purpose or the manner in which the data will be processed and are thus not considered a data controller for this agreement.

LEGAL BASIS FOR PROCESSING DATA:

- GDPR (personal identifiable data): Article 6(1)(e) - Processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller. As a higher education establishment, the University conduct research to improve health care and services.

- GDPR (special category data): This data is processed on the additional condition of: Article 9(2)(j) – Processing is necessary for archiving purposes in the public interest, scientific or historical research purposes.

METHODOLOGY

1. Keele University will send in a Cohort of 381 consented patients Study IDs plus NHS Number, Forename, Surname, and Date of Birth to NHS England via a Secure Electronic File Transfer (SEFT) service.

2. NHS England will link the cohort and extract the HES and Mortality data, and then remove the identifiers. NHS England will then send the pseudonymised extracts back to Keele University via SEFT.

3. Keele University will link the pseudonymised data from NHS England* to pseudonymised data from the UK Renal Registry and pseudonymised BISTRO trial data (e.g. Case Report Forms and Questionnaires) via the Study ID to form the final pseudonymised record-level BISTRO trial data set.

*Similar data linkage requests are being processed by the equivalent organisations in the devolved nations.

4. Keele University will then share the final record-level pseudonymised BISTRO trial data set with the University of Warwick for joint data processing and analysis.

Reporting and export procedures for data downloads to common statistical software analysis packages (SPSS, SAS, Stata, R) are provided within Keele University’s secure network.

For the purpose of identifying the correct cohort and obtaining agreed pseudonymised data linkage for the trial the appropriate variables will be shared with UK Renal Registry (renal data, UKRR) and NHS England- equivalent bodies in devolved nations (for equivalent HES and mortality data). No special category data, specifically no health data, will be sent from Keele University. UKRR and NHS England- equivalent bodies in devolved nations will return a pseudonymised dataset (BISTRO BISTRO Study ID and variables containing no identifying data) to Keele University for data linkage.

DOWNLOADING DATA FROM SEFT AND UPLOAD TO KEELE CTU SERVER

A member of the BISTRO study team will request Keele University IT to create a temporary folder with restricted access on the secure Keele University server (logging this action in the audit history). The data recipient will then use a Keele University secure remote device using 2-factor authentication via VPN to log into NHS England’s Secure Electronic File Transfer (SEFT) system and save downloaded file(s) into the temporary folder.

After the data download, the data recipient – who is still accessing the server via a secure remote device will upload the NHS England data from the temporary folder into the Clinical Trials Unit (CTU) secure server space as soon as possible with a 2-factor authentication via Virtual Private Network (VPN) enabled for each team member.

A member of the BISTRO study team will request Keele University IT to destroy the initial download folder with the appropriate software and ensure this is logged on the audit history.

NHS England data linkage and analysis will be performed on the Keele University CTU server. Keele University is able to limit access within the CTU server according to department and permissions. In this instance, the BISTRO study area in Password access is randomised and changes every 60 days.

BISTRO Trial identifiable data is stored separately to the pseudonymised datasets, and no linkage will be attempted.

Notes:

- The use of Microsoft Office 360 is for a transient, short-term data transfer method only. Data is not stored on SharePoint for longer than the period of transfer and no data analysis/manipulation will be conducted on this Cloud Platform.

- All Keele and Warwick substantive staff complete mandatory Information Governance and GDPR training on an annual basis. Please note that Warwick will only have access to NHS England data after the data linkage has taken place at Keele University.

- Physical access to the both University’s offices are restricted to personnel authorised to access these premises. The computers used are encrypted or work within an encrypted environment.

The Keele University will store the identifying details of participants linked with a pseudonym BISTRO Study ID and will separately store the “clinical” data (i.e. data provided by participants, recruiting sites and obtained from other providers such as NHS England UKRR, etc.) as “pseudonymised” data containing the pseudonym BISTRO Study ID but no other identifying details.

The two datasets (initial PID containing dataset submitted to NHS England and returned pseudonymised dataset) would be kept separate and not re-linked. Access to the database with identifiable information will be restricted according to the Keele University’s Data Security and Protection Toolkit. Datasets from NHS England (and UKRR) would be added into the trial database using the pseudonym BISTRO Study ID for linkage with other data. For patients in England, this includes data from: the trial case report forms, participant questionnaires and UKRR.

Appropriate pseudonymised variables from HES and Civil Registration data will be shared with the University of Warwick. It will be used to derive event rates (e.g. hospital admission rates and death rates) for Health Economics analysis, together with BISTRO Study ID (pseudonymised dataset) will be shared with University of Warwick (data processor). A separate data sharing agreement has been established with the data processing organisation.

Microsoft Limited provide Office 360 SharePoint Cloud Services for the University of Warwick and Keele University and are therefore listed as a data processor. They supply support to the system, but do not access data. Therefore, any access to the data held under this agreement would be considered a breach of the agreement. This includes granting of access to the database[s] containing the data.

TRANSFER OF DATA FROM / TO WARWICK UNIVERSITY FROM KEELE UNIVERSITY

As a part of the data analysis, a pseudonymised subset of the trial data (using the BISTRO Study ID), with pre-defined appropriate variables, will be then shared with University of Warwick via SharePoint (Microsoft Office 365) which both Universities have.

The files are in the form of excel documents with password protection applied.

A member of the BISTRO study team will request Keele University IT to create a temporary folder with restricted access on the secure Keele University server (logging this action in the audit history).

A member of the BISTRO study team will move the files from the Keele CTU Server, to the temporary folder, and then upload the file to a specifically created secure folder on Microsoft Office 360 SharePoint. SharePoint access to the folder requires a 2-factor authentication for the folder to be opened, i.e. authorised person’s email address and a random code generated every time access is required. The folder will be accessible to only authorised University of Warwick substantive staff that will be performing the analyses. In order to access the folder, the users will be sent a link to the folder and will need to request a code to enter the folder. This code will need to requested every time access is required. The code will be different for each request.

A member of the BISTRO study team will request Keele University IT to destroy the initial download folder with the appropriate software and ensure this is logged on the audit history.

A member of the BISTRO study team at the University of Warwick accesses the secure folder on Microsoft Office 360 SharePoint and downloads the files directly into the Warwick University’s CTU secure server. Study team members at University of Keele and University of Warwick must ensure the data files within the Microsoft 360 SharePoint folder is then deleted and removed from the recycle bin.

The data will be processed and analysed in the Warwick CTU secure server.

When all the analyses have been performed, the pseudonymised data will be shared back to Keele University via SharePoint (as described in steps 6 – 1 above in reverse)

Trial data will be processed at specific trial time points: upon receipt for data linkage, data analysis including the use of the datasets by external processors (e.g. University of Warwick).

HES and Civil Registration data will be processed to derive the secondary outcomes for the trial, as set out above. These will then be added to the main analysis file for the trial. Both statistical and health economic analysis plans have been developed and were signed off by the trial steering committee. This will be followed to address secondary objectives of the trial to determine the effect of the intervention on:

- Significant events, including vascular access failure and associated interventions, cardiovascular events, hospital admissions and death, including the use of routinely collected data and long-term legacy effects beyond trial completion using data linkage to routine health care databases such as the UK Renal Registry (via the Renal Registry), Hospital Episode Statistics and Mortality data via NHS England (and their equivalent bodies in Wales, Scotland and Northern Ireland).

- Objective measures of dialysis efficacy and safety: e.g. inter-dialytic fluid gains, intra-dialytic hypotension, urea-reduction ratios (routine data)

- Cost effectiveness of the intervention – a cost-utility analysis will be undertaken to determine the primary outcome of incremental cost (or cost savings) per quality-adjusted life year (QALY) gained.

DATA RETENTION

Retention of data at participating sites (site archiving) will be for 5 years from the end of the trail - this is only applicable for BISTRO Trial collected data by the site, for their participants, not the NHS England data referred to in this agreement. At the end of the 5 year period, each site will be contacted to destroy this data. The reason for this is to resolve any potential queries after data lock.

Retention of the final dataset for the BISTRO trial will be for 10 years post end of the study as per Keele Standard Operating procedure (SOP 17 - CTU Archiving and Destruction) at the Keele University approved location. This will include the data linkage data, e.g. pseudonymised NHS England data.

HES DISCLOSURE CONTROL / SMALL NUMBER SUPPRESSION

In order to protect patient confidentiality, when presenting results calculated from HES record level data, outputs will contain only aggregate level data with small numbers suppressed in line with HES Analysis Guide. When publishing HES data, you must make sure that:

· cell values from 1 to 7 are suppressed at a local level to prevent possible identification of individuals from small counts within the table.

· Zeros (0) do not need to be suppressed.

· All other counts will be rounded to the nearest 5.

Data will not be made available to any third parties other than those specified except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

The trial steering committee has approved the dissemination strategy, with approval by the Sponsor (Keele University). This was developed in close collaboration with the trial management group, Advisory and Dissemination Board facilitated by Kidney Research UK (on behalf of the UK Kidney Research Consortium), BISTRO Patient Advisory Group (PAG) and the trial steering committee.

Trial progress and results aim to be disseminated by other communication channels at Keele University, such as Keele Clinical Trials Unit website, Keele University social media account, and Keele staff and public news (weekly publications) as and when relevant. As BISTRO is funded by NIHR HTA, the trial follows the NIHR research outputs and publications guidance: https://www.nihr.ac.uk/documents/nihr-research-outputs-and-publications-guidance/12250#Notifying_NIHR_of_upcoming_research_outputs

Trial progress and results are hoped to be disseminated to patient organisations, e.g. Devices for Dignity, with the support of the BISTRO PAG.

An BISTRO social medial twitter account (@Keele_BISTRO) and a trial specific website (https://www.keele.ac.uk/bistro/) have also been set up to keep interested patients, carers, clinicians, managers and policy makers up-to-date with trial progress.

Peer-reviewed publications - Data will only be published in aggregated format with small number suppression applied as per the HES analysis guide.

The BISTRO trial has a dissemination plan - the following outputs are planned (estimated 2022/2023):

• Economic evaluation of the intervention and analysis of the benefits of residual kidney function

• In-depth analysis of effect of the intervention on the patient activation and engagement with fluid management

• The impact of dialysis unit practice patterns on primary and secondary endpoints

• Publication of the clinically validated fluid assessment tool and associated educational material, with the potential to develop this into an application suitable for hand-held devices.

Routes of dissemination will be as follows:

• National Meetings (to include study updates and findings):

• Renal Association and British Renal Society (e.g. annual Kidney Week, usually multidisciplinary meeting covering all aspects of nephrology and dialysis treatment).

• International Meetings (via submitted abstracts):

• American Society of Nephrology, European Dialysis and Transplantation Association/European Renal Association, European Dialysis and Transplantation Nurses Association, International Society of Nephrology.

Abstracts for poster and/ or oral presentation aim to be submitted to national and international conferences, such as UK Kidney Week, The European Renal Association – European Dialysis and Transplant Association (ERA-EDTA), American Society of Nephrology and World Congress of Nephrology (estimated 2022/2023).

The funder (NIHR) requires sight of all output before it is made public (with the exception of individual tweets).

The expected measurable benefits from this research are not expected to be realised for health care until the main results of the trial are published in 2022.

The BISTRO trial has been cited in the National Institute for Health and Care Excellence (NICE) NICE guidance: “Multiple frequency bioimpedance devices to guide fluid management in people with chronic kidney disease having dialysis - Diagnostics guidance [DG29]” published on 21st June 2017 (https://www.nice.org.uk/guidance/dg29) . Its recommendation states:

“There is currently not enough evidence to recommend the routine adoption of the BCM – Body Composition Monitor to guide fluid management in people with chronic kidney disease having dialysis in the NHS. Further research is recommended to show the effect of using the BCM – Body Composition Monitor on clinical outcomes.

Centres that are currently using the BCM – Body Composition Monitor to guide fluid management are encouraged to take part in research and data collection. Centres that do not currently use the BCM – Body Composition Monitor to guide fluid management should only do so as part of a research study, such as the BISTRO trial.”

The publication of BISTRO trials results is hoped to partially inform the future review of NICE guidance DG 29.

SPECIFIC BENEFITS TO PATIENTS AND THE HEALTHCARE SYSTEM

The research findings aim to inform guidance on the clinical value of using a bioimpedance device to guide fluid management in haemodialysis patients. As soon as the results of the study are peer reviewed and published they will be submitted to NICE so as to inform the next iteration of guidance on the use of this technology. In the previous iteration “Multiple frequency bioimpedance devices to guide fluid management in people with chronic kidney disease having dialysis - Diagnostics guidance [DG29]” published on 21st June 2017 (https://www.nice.org.uk/guidance/dg29) NICE could not reach a conclusion as to whether the use of this technology could be supported, stating explicitly in their report that they would not review the evidence again until BISTRO had reported. Key to this assessment by NICE, which will be undertaken by third parties, will be the clinical benefits if found (prolongation of residual kidney function which improves wellbeing and survival of people on dialysis) and an understanding of the cost of this intervention (equipment, training) and potential efficiency savings (e.g. less requirement for dialysis, fewer complications and hospital episodes).

The guidance, likely completed in 2022, aims to be generalisable to all dialysis units in the UK (72) which currently treat 25,000 people, approximately 5,000 commencing treatment each year. Following the recommendation from NICE uptake of the guidance will be audited nationally via the UK Renal Registry or equivalent process, such as GIRFT or reporting to the local dialysis networks (currently being redesigned as part of a national re-organisation of renal services)."

OVERARCHING BENEFITS TO PATIENTS

The maintenance of residual kidney function in patients commencing dialysis is associated with considerable advantages, not least improved patient survival. Peritoneal dialysis is a treatment for kidney failure. It uses the lining of your abdomen, or tummy, to filter your blood inside your body. This filtering will remove waste products that have built up in your blood stream. The Canada-USA (CANUSA ) study found that each 250 ml of urine per day increased survival by 36% in peritoneal dialysis patients. In addition, in the NECOSAD study (an analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD )-2) complete anuria, when the kidneys stop to produce urine, in haemodialysis patients increased the relative risk of death 17 fold compared to those with relatively well preserved kidney function (still only a few percentage points of normal).

Other potential benefits include improved wellbeing, better quality of life and less need to remove high fluid volumes during dialysis sessions with its attendant risks of intra-dialytic hypotension, cardiac stunning and potentially increased mortality. It is therefore surprising how few clinical trials have focussed on interventions to maintain residual kidney function as a key benefit to patients – the exception being ultrapure water which is now standard care. Worse than this, a frequently applied fluid management strategy is to reduce the post-dialysis target weight until minimal or no anti-hypertensive drugs are required as evidence that adequate control of volume status is achieved. BISTRO’s recent survey of fluid management practice patterns in UK units indicate that this is still being pursued in the majority of units despite the risk it poses to residual kidney function by setting in place a continuing vicious cycle of volume depletion, excessive thirst and high inter-dialytic weight gains. Once anuria (complete loss of kidney function) has developed then patients are severely restricted in their daily fluid intake and dietary restriction, which in turn exacerbates their poor nutritional status. The introduction of bioimpedance technology provides clinicians with the opportunity to potentially make more accurate decisions to help to control the volume of fluid left in the body after haemodialysis. The potential benefit to patients would be a change in clinical practice, where clinicians can use this to guide how much fluid should be removed from the body with the normal clinical assessment of the amount of fluid in the body; that is associated with improved wellbeing, fewer dialysis related symptoms, possibly less dialysis in those commencing treatment in an incremental fashion and potentially better survival.

OVERARCHING BENEFITS TO THE NHS

Of 54,000 people in the UK treated with kidney replacement therapies (accounting for ~2% of the total NHS budget), 24,000 receive centre-based haemodialysis at an annual tariff of £24,000 excluding additional costs such as travel, drugs, access procedures and inpatient episodes. In this high-cost setting bioimpedance has the potential to enhance the productivity of haemodialysis care by helping clinicians make appropriate and safe treatment decisions as defined by principles underpinning the Department of Health’s Quality, Innovation, Productivity and Prevention (QIPP) Policy. BI also has potential to address several of the NHS Outcomes Framework domains, including prevention of premature death, improving outcomes by addressing a number of NICE chronic kidney disease standards such as cardiovascular risk, blood pressure and avoidance of acute illness episodes and enhancing the quality of life for people on dialysis (i.e. long-term condition), and contributing through improved engagement and activation to a more positive patient experience.

YIELDED BENEFITS PROVIDED BY APPLICANT 02/08/2024

The NHS England data on trial participants' use of NHS services (admissions, procedures etc) was used to determine the health care costs incurred whilst participating in the BISTRO trial. They were combined with Quality of Life (QoL) measurements (using EQ-5D-5L and SF-12 instruments) to determine the cost utility of using a bioimpedance device to improve the accuracy of making fluid assessments when compared with a proforma alone in dialysis patients. Even though the main trial outcome (loss of residual kidney function) was not affected by the addition of bioimpedance measurements, bioimpedance use was associated with modestly reduced costs and slightly better maintained QoL. Overall it was calculated that there was a 74% chance that bioimpedance was cost effective. This finding will be used by NICE technology evaluation to provide guidance on the use of bioimpedance in dialysis patients. The study team will now use the same data to establish whether the preservation of residual kidney function was associated with a cost saving.