Improving the safety and continuity of medicines management at care transitions (ISCOMAT): a cluster Randomised Controlled Trial

Bradford Teaching Hospitals NHS Foundation Trust and the University of Leeds are requesting to use NHS Digital data for a study entitled "Improving the Safety and Continuity Of Medicines management at Transitions of care" (ISCOMAT).

ISCOMAT has received a favourable ethical opinion, is funded by the National Institute for Health Research (NIHR) and is performed by a public authority. Bradford Teaching Hospitals NHS Foundation Trust's and the University of Leeds' lawful bases for processing personal data and health data (defined as special category data) for the purposes of this project are:

Article 6:1(e): Specific task in the ‘public interest’ or task that has a clear basis in law, and

Article 9:2(j): Special category data used for “Archiving in the public interest, scientific or historical research or statistical purposes”, with a basis in law.

Participants consented to have their personal data shared with NHS Digital, including personal details (initials, date of birth, postcode and NHS number) to be shared with NHS Digital for this project. Participants who withdrew their consent for data collection from routine sources will not form part of this cohort for this data application. All consent documentation and subsequent amendments to it were reviewed and approved by the ISCOMAT Patient Led Steering Group.

ISCOMAT Background and rationale:

Making the use of medicines as safe and effective as possible are priorities for patients and healthcare providers. When a patient moves, for example from hospital to home, medicine problems are common and planned changes are not always followed through. Patients particularly at risk are those with long-term illnesses taking several medicines – especially when medicines have been started or changed in hospital.

Heart failure can be managed through a combination of several different drug combinations at titrated doses. Successful management of heart failure patients’ medication can reduce hospitalisations, improve quality of life and increase survival rate. Optimising the management of medication as patients transition from hospital to home is complex, and requires co-ordinated input from different healthcare organisations, including hospitals, pharmacies and general practice, and different healthcare professions.

Heart failure affects 26 million people globally, approximately 900,000 people in the UK and the incidence of new diagnoses of the condition is increasing, affecting older people and those from lower socio-economic groups in greater numbers.

Aim:

ISCOMAT is a programme of interlinked work packages, which developed a Medicines at Transitions Intervention (MaTI) toolkit to help the way heart failure patients are supported with their medicines. MaTI was developed through a process of co-design with patients and healthcare staff, with the aim to improve the safety and continuity of medicines at care transitions, and subsequently reduce mortality and readmission to hospital.

The MaTI toolkit will be evaluated in the ISCOMAT Trial. This is a cluster randomised controlled trial (cRCT), which randomized 43 secondary care cardiology services (clusters) to implement either the MaTI or continue to deliver their usual care.

Cohort:

A total of 1615 men and women aged over 18, who were admitted to a participating hospital with a diagnosis of heart failure and evidence of at least moderate left ventricular systolic dysfunction confirmed within the last 5 years, were recruited between 06/2018 and 03/2020 and consented to share their information with NHS Digital.

Outcome and data required:

The main (primary) outcome is all-cause mortality and Heart Failure related hospitalisation within 12 months of recruitment. This is in line with the aim of the MaTI intervention, which is to reduce mortality and reduce readmissions to hospital.

The key secondary outcome is continued prescription of guideline indicated therapy at 12 months post recruitment. This will help identify if the MaTI toolkit increases the number of heart failure patients who are receiving appropriate heart failure medication, after they are discharged from hospital to home.

Other secondary outcomes will help identify any other potential benefits of the MaTI toolkit, and include: time from registration to death, or to heart failure rehospitalisation; length of time on guideline recommended medications from registration, patient understanding of medicines and satisfaction with medicines related care; number of days alive and out of hospital; hospitalisations; death due to specific causes.

A parallel cost-effectiveness analysis will include Hospital Episode Statistics (HES) Admitted Patient Care (APC), HES Critical Care (CC), HES Outpatients, Emergency Care Data Set (ECDS), Civil Registration (Deaths) Secondary Care Cut (CRD SCC) to derive a costs per all-cause deaths prevented and re-hospitalisation prevented, alongside cost-effectiveness of intervention.

Data to support primary and secondary outcomes as outlined above is available in HES, ECDS, CRD SCC, and the Medicines Dispensed in Primary Care (NHSBA) data.

Data will also be sought from the National Heart Failure Audit via the National Institute of Cardiovascular Outcomes Research (NICOR), clinical data directly from participating hospitals, and patient reported outcomes direct from participants.

Patient-level data is needed from HES, ECDS, linked Civil Registration Mortality Data and Prescribing data to allow the trial team to link these data with the NICOR data, clinical data and patient reported outcomes for our cohort of participants. The linked datasets will subsequently be used as the analysis dataset for the ISCOMAT trial.

Data is requested from the time the first consenting and eligible participant was admitted to hospital (June 2018) until 12 months after the last participant was registered (March 2021). These years are required in order to achieve the outcomes listed above. The hospitals taking part in ISCOMAT are spread throughout England and therefore necessitates the geographical spread of data requested.

Obtaining data held in electronic health records will provide a complete dataset to answer the primary and secondary outcomes. The trial was designed to use data from routine sources as it was determined that this would be the least intrusive way to answer research questions for participants and to also minimise burden on research staff at hospitals.

Data requested for this application is minimised as much as possible. This data sharing agreement relates to the cohort of 1615 (1615 out of 1641 recruited consented to share data with NHS Digital) men and women aged 18 and over at time of admission, with a diagnosis of heart failure with evidence of at least moderate left ventricular systolic dysfunction confirmed within the last 5 years. The specific data requests relate to this cohort and any admissions to hospital and instances of death, including cause of death between the date of registration and 12 months. Obtaining prescribing datasets also supports key secondary outcomes on length of time participants are prescribed guideline indicated therapies. The data requested is proportionate to the data required to answer the primary and secondary outcomes.

Organisations involved:

Bradford Teaching Hospitals NHS Foundation Trust (BTHFT) and the University of Leeds (UoL) are joint controllers. The data is processed at the University of Leeds.

The Chief Investigator is a professor of cardiovascular medicine and honorary Consultant Cardiologist. His substantive employer is the University of Leeds, and he has an honorary contract with BTHFT. The Clinical Trials Research Unit (CTRU), at the University of Leeds, is co-ordinating the trial and is responsible for trial design and conduct, including protocol development, ethics approval, trial monitoring, data collection and ongoing management, and statistical analysis and reporting. All data for the cohort of patients recruited to the trial is processed and stored at the CTRU. The Academic Unit of Health Economics (AUHE) at the University of Leeds will perform the cost effectiveness analysis, and data will be transferred to the trial health economist for processing. The CTRU and the AUHE have both completed the Data Security and Protection Toolkit (DSPT) which evidences each department's compliance against the National Data Guardian’s 10 data security standards.

ISCOMAT is delivered by a multidisciplinary research collaboration between BTHFT (as NHS lead and Sponsor), the University of Bradford and University of Leeds. The collaboration includes patient representatives, and a Patient Led Steering Group is actively involved in conduct and oversight. A contract is in place between the funder (Department of Health) and BTHFT (NHS Lead), and a collaboration agreement is in place between the BTHFT, the University of Leeds, and the University of Bradford. The ISCOMAT trial is the fifth in a series of interlinked work packages that comprise a NIHR Programme Grant for Applied Research (PGfAR). An earlier work package (1B) successfully obtained data from NHS Digital for a cohort of 53 participants (DARS-NIC-40493-G5Y6K). The University of Bradford are not receiving, or otherwise accessing and processing, any data under this data sharing agreement. Processing activities are undertaken at the University of Leeds as fully outlined in the Processing Activities section.

Decisions relating to the processing of personal data remain with BTHFT and UoL. University of Leeds and BTHFT as joint controllers confirm that they will ensure a GDPR compliant, publicly accessible transparency notice is maintained throughout the life of this agreement.

The trial is funded by the NIHR Programme Grants for Applied Research (RP-PG-0514-20009). The trial funder does not make decision on the processing of personal data, and has no role in study design, data collection, data analysis, data interpretation, or writing of the final report.

The trial statistician at the Clinical Trials Research Unit (CTRU) will provide the following identifiers: sex, date of birth, postcode and NHS number. Study ID for each participant will be provided to allow linkage of the NHS Digital data to the trial data held by the CTRU. For each participant, date of admission to hospital, and 12 month post registration will be provided. This will provide assurance that minimum data required is obtained. The data will be transferred to NHS Digital via the required secure platform.

NHS Digital will perform an automatic cohort tracing system to search the data provided against existing electronic records to generate a set of records that belong the cohort of 1615 trial participants. This will include linkage of data from Hospital Episode Statistics (HES) Admitted Patient Care (APC), HES Critical Care (CC), HES Outpatients, Emergency Care Data Set (ECDS), Civil Registration (Deaths) Secondary Care Cut, and the Medicines Dispensed in Primary Care (NHSBA) data.

Opt-outs will not be applied, as all participants included in the cohort have consented for their personal data to be shared with NHS Digital to obtain information from their electronic health records, and have not withdrawn their consent for sharing of their personal data with NHS Digital for this purpose.

NHS Digital will transfer the linked record level data to the trial statistician on one occasion. NHS Digital will return pseudonymised data linked using the CTRU supplied Study ID. On receipt of data from NHS Digital, the trial statistician will transfer the data into a secure folder prior to using in statistical software for analysis. As a first step, the trial statistician will perform data cleaning in accordance with CTRU Standard Operating Processes, for example, to ensure no duplicate episodes, no admission after date of death (if deceased). The trial statistician will subsequently link the data obtained from NHS Digital with the trial data held at the CTRU on this cohort, to derive a trial analysis dataset. Linkage of the data will be by the Study ID. Participants in the trial dataset will only be identifiable by their Study ID and all statistical analysis will be conducted with reference to this – there will be no link to patient identifiable data, all data will be pseudonymised.

The statistical analysis will be conducted within a statistical software package in accordance with a Statistical Analysis Plan which will be finalised and agreed by the Trial Management Group in advance of any analysis being undertaken. For the analysis of the primary outcome of all-cause mortality and heart failure rehospitalisation over 12-months follow-up, an indicator will be derived for each participant based on NHS Digital data to indicate whether or not the participant has experienced the primary outcome. The outcome will be compared between the trial arms using a random effects logistic regression model which will include trial data held by CTRU. A similar approach will be taken for the analysis of the secondary outcomes with the most appropriate statistical model chosen based on the type of outcome.

Data items specified in the Health Economics Analysis Plan will be transferred to the trial health economist, via CTRUs Secure File Transfer. Trial statisticians and health economists who access the data are all substantive employees of the University of Leeds, and undergo relevant Data Protection and Information Security training on a regular basis as part of their employment at the University of Leeds.

All data for the cohort of patients recruited to the trial is processed and stored at the CTRU (Organisation Data Service (ODS) is ECC0010), and the Academic Unit of Health Economics (AUHE) at the University of Leeds will perform the cost effectiveness analysis (ODS is 8E218-SEED). No record level data disseminated via NHS Digital for this application will flow from the University of Leeds.

The CTRU server infrastructure is split between two data centres on the main University of Leeds campus. Backups taken from this infrastructure are replicated to the University of Leeds disaster recovery site at the University of York, with tape backups encrypted to AES 256 being kept at Iron Mountain. In all locations data is stored encrypted on disk/ tape. The ability to decrypt is only held by the CTRU. The two disaster recovery sites provide different recovery options. The site at University of York is a warm online copy on disk that can be retrieved instantly for the last 30 days. The Iron Mountain site holds a cold offline copy on tape for 12 months. Neither site can access the data. At the University of York location, the server is University of Leeds own equipment and connects only to the University of Leeds network. This server holds the disaster recovery copy of University of Leeds backups. There is no data processing at this site. The University of York only provides the physical space and power to support this.

Only summary results (no record-level data) will be made publicly available via open access publications, including the NIHR Journals Library publication of the final report. No third party will have access to the data. Data will not be used for any other purposes, and it will not be used for commercial purposes, nor for direct marketing purposes.

The data will only be accessed by authorised members of the team who are all substantive employees of the data controllers through a secure drive on the University of Leeds network. The files will be restricted to those processing data.

The main analysis is due to be completed by March 2022, with the aim to publish main trial results in late spring, early summer 2022. Results will be published in open access and peer reviewed journals, including publication via the National Institute for Health Research's own journal library, the Lancet, the British Medical Journal (BMJ) and the European Heart Journal (EHJ). Results will also be presented at relevant conferences linked to cardiology and patient safety, including the European Society of Cardiology Congress (Summer 2022). Longstanding and ongoing engagement with stakeholders, including both scientific and policy-making audiences will provide a direct pathway to impact for the outputs of this research.

The Patient-Led Steering Group will inform the dissemination strategy and its members will play an active role in the format and content of academic papers (specifically patient implications) and will present at local, regional and national conferences and wider stakeholder meetings.

Only analyses featuring aggregated data with small number suppression will appear in outputs.

The results and outputs of the study will be further communicated via the study team's websites, social media accounts and through other public promotion of research utilising the study team’s networks, including clinical networks, scientific networks and charitable organisations with heart failure and medicines safety focus.

Lay summaries will be added to the trial website (https://www.bradford.ac.uk/iscomat/) and the trial registry will be updated (https://doi.org/10.1186/ISRCTN66212970). Summaries will also be provided to participating sites. The Use My Data citation - “This work uses data provided by patients and collected by the NHS as part of their care and support” - will be used in all outputs as appropriate, to further acknowledge that the research involved patient data. These outputs will follow the main analysis in 2022.

If the intervention is shown to be effective (as supported by NHS Digital data linkage for primary outcome), the network of champions (patients, staff and other stakeholders) created during the programme will be used to spearhead a plan for adoption and spread. Materials developed during the programme will be made available for wider use.

Heart failure affects 26 million people globally and approximately 900,000 people in the UK. With the incidence of new diagnoses of the condition increasing, the ISCOMAT trial results have the potential to inform the treatment and care of heart failure patients when especially vulnerable during a care transition. The benefits from this dissemination will not be realised for health care until the main results of the trial are published in 2022.

The benefits focus on patient care improvement, and dissemination will be led by the trial research team, including programme management group, trial management group and trial steering committee. The results, placed in open-access peer reviewed publications, are hoped will provide an evidence base with potential to impact on clinical guidance, including National Institute for Health and Care Excellence (NICE) guidelines on heart failure and medicines related care. It is hoped that benefits for health service commissioners will include provision of evidence to support future commissioning of community pharmacy services as well as the need for more effective optimisation of heart failure treatment, with associated health benefits. The aim of this is to ensure that patient medicines management is optimised, and the burden of cardiovascular disease is reduced through preventable cardiovascular events that occur in the period after patients with heart failure are discharged from hospital. Our findings are intended to be, with modifications, transferable to other patient groups who have long-term conditions, frequent hospital readmissions and polypharmacy.

Heart failure affects 26 million people globally and approximately 900,000 people in the UK. With the incidence of new diagnoses of the condition increasing, the ISCOMAT trial results have the potential to inform the treatment and care of heart failure patients when especially vulnerable during a care transition. The benefits from this dissemination will not be realised for health care until the main results of the trial are published in late 2024.

The benefits focus on patient care improvement, and dissemination will be led by the trial research team, including programme management group, trial management group and trial steering committee. The results, placed in open-access peer reviewed publications, are hoped will provide an evidence base with potential to impact on clinical guidance, including National Institute for Health and Care Excellence (NICE) guidelines on heart failure and medicines related care. It is hoped that benefits for health service commissioners will include provision of evidence to support future commissioning of community pharmacy services as well as the need for more effective optimisation of heart failure treatment, with associated health benefits. The aim of this is to ensure that patient medicines management is optimised, and the burden of cardiovascular disease is reduced through preventable cardiovascular events that occur in the period after patients with heart failure are discharged from hospital. Our findings are intended to be, with modifications, transferable to other patient groups who have long-term conditions, frequent hospital readmissions and polypharmacy.