PATTErn: A study of Physical Activity paTTerns and major health Events in older people with implantable cardiac devices

Aim/Purpose

Data from NHS Digital is requested as part of the PATTErn study (A study of Physical Activity paTTerns and major health Events in older people with implantable cardiac devices) to provide record-level information about healthcare utilization for each participant.

The purpose of the PATTErn study is to examine the relationship between physical activity (as measured by cardiac devices) and non-elective hospitalisation attendances/admissions (NEHA) in older people with cardiac devices. The hypothesis is that physical activity tends to decline in the days/weeks leading up to NEHAs. The study will be exploring physical activity trends surrounding NEHAs, thus require information about these events in order to perform the analysis. Patients are being recruited from a tertiary care service; therefore, NHS Digital HES data is required as many participants’ local hospital will be elsewhere in the region.

Article 6 GDPR justification for processing

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’

All patients participating in the PATTErn are required to provide written, informed consent. This is recorded on the HRA approved consent form. Two sections on the consent for refer to NHS Digital data:

1) ‘I confirm that I have read and understand the Participant Information Sheet dated 13/08/2018 (version 4.4) for the above study and have had the opportunity to consider the information’.

- The patient information sheet contains detailed information about collection and use of data from NHS Digital (attached)

2) ‘I understand that information about my hospital attendances in the last 12 months will be collected from NHS Digital. This process will involve sharing of personal details with the NHS Digital service.’

The collection of data from NHS Digital is essential for the completion of the PATTErn study. This is in the public interest as: (1) all participants have expressed willingness of their data to be used as part of the PATTErn study to advance academic knowledge (as evidenced above), (2) results from the PATTErn study will advance academic and clinical knowledge and understanding of the relationship between PA and hospitalisation events and (3) results from the PATTErn study will in likelihood lead to either further clinical studies or direct change in clinical service provision with the aim of improving patient care.

Article 9 GDPR justification for processing

9.2(j) processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

As laid out above, data processing is in the public interest for the benefit of health and social care. Safeguards to protect data subjects are in place as per University of Manchester data security and data protection policies (see processing activities).

How will the data requested achieve the aim identified?

NEHA data from NHS Digital will allow the study to investigate the relationship between physical activity and non-elective hospital attendances/admission. This is the core aim of the PATTErn study.

NHS Digital data will provide dates and information about NEHAs. NEHA ‘events’ will be added to participant timelines to facilitate temporal trend analyses.

Background to Study

The PATTErn study is a stand-alone study. It is sponsored by the University of Manchester, and forms the central part of a PhD.

PATTErn is an observational research study with full HRA approval (granted 18th June 2018, substantial amendment approval granted 1st October 2018) which opened for recruitment on the 12th October 2018.

the Phd is being carried out by a substantive employee of the University of Manchester.

The PATTErn study is a stand-alone study.

It is not part of a wider project or collaboration. No follow-up work is currently planned. NHS Digital data collected as part of this application will be used as described above for the PATTErn study only.

Data Subjects

Recruitment target is 150 participants (it is anticipated recruitment will extend above this target). There is a single cohort. Inclusion criteria identify patients over the age of 60 under follow-up at Manchester Heart Centre with a Medtronic cardiac device in situ for > 6 months, which has the capacity to monitor physical activity. Patients must be able to provide written, informed consent in the English language.

Purpose of Request

The purpose of this project is to investigate physical activity patterns in older people with cardiac devices, and how these correlate with non-elective hospitalisation events.

I. Data regarding (1) non-elective hospital attendances and (2) non-elective hospital events is required to provide the ‘outcome’ data for this study.

II. Information regarding the dates of these events is required to facilitate predictive modelling.

III. Data is required to be identifiable so we can link events with each study participant.

IV. Data for the 12 months prior to date of recruitment is required as per study protocol and patient information sheets/consent forms.

V. There are no less intrusive ways of achieving this data reliably. It is not feasible to rely on patients to remember dates of hospital attendances.

Organisations involved in study

1. University of Manchester (sole data controller and data processor)

Role: sponsor of study

Data remit: facilitate transfer of data from NHS Digital, storage, processing and analysis of data

2. Manchester University NHS Foundation Trust (MFT) - **note previously known as Central Manchester NHS Foundation Trust**

Role: study site

Data remit: MFT will not receive any NHS Digital data. It will, however, provide study participant identifiers to NHS Digital, as this is where personal data from study participants is collected and stored, including the enrolment log, which links personal information with the study participant number (SPN). MFT will not have access to disseminated NHS Digital data.

3. Medtronic Inc

Role: funding the post of Principle Investigator and some research costs associated with the study (from April 2018-April 2019)

Data remit: Medtronic have no access to any of the NHS Digital data. It is not involved in this aspect of the study. Medtronic's role in the study is to facilitate processing of pseudo anonymised device downloads.

4. British Heart Foundation

Role: funding the post of Principle Investigator (BHF Clinical Research Training Fellowship) and some research costs associated with the study (from April 2019 – April 2021)

Data remit: The BHF have NO ACCESS to identifiable data or NHS Digital data. It is not involved in this aspect of the study. Will assist in dissemination of results.

The PATTErn study is not being undertaken for commercial purposes, however there is commercial involvement in the study. The PATTErn study is an academic study, forming part of the Principal Investigator’s (PI) PhD. The PI's post at Manchester University is funded by Medtronic - a company which manufactures cardiac devices. Funding from Medtronic ended in April 2019 when funding was taken over by the British Heart Foundation.

Funders have no influence over study results or outputs generated. Funders have no access to NHS Digital data.

Sharing of identifiable data from participants with NHS Digital is included in the Participant Information Sheet (which all participants must confirm they have read before consenting to participate), and is explicitly stated (with corresponding participant initials) on the consent form.

Data Flows

Information transferred to and from NHS Digital will be record-level, directly identifiable patient data.

Personal data for each study participant will be provided to NHS Digital by the study site - Manchester University NHS Foundation Trust (MFT). This data will be Study ID (unique 6-digit number) and personal identifiers only: NHS Number, Date Of Birth, Gender, Postcode.

NHS Digital data will then be provided to the study sponsor - the University of Manchester. Data will be received directly into the Data Safe Haven (see below). The output of NHS Digital data received will be record-level. Data will be tagged with study participant number (SPN), and personal identifiers (NHS Number, Date of Birth, Gender, Postcode). Disseminated data from NHS Digital will not be shared with any other organisation (including MFT and Medtronic Inc).

Medtronic Inc role is the funding the post of Principle Investigator and some research costs associated with the study they have a remit to facilitate processing of pseudonymised device downloads. No access to identifiable data or NHS Digital data is permitted the information is limited to that of the device information.

The Data Safe Haven (DSH)

NHS Digital data will be received via the Data Safe Haven (DSH) at the University of Manchester (UoM). Access to the DSH is only granted to substantive employees of the UoM whom have completed the University’s Data Protection Training and authorised by the Research Governance, Ethics and Integrity team (RGEIT). Each DSH user has a project file within the DSH, which is accessible to that user only.

Access to the Data Safe Haven is restricted to on-campus connectivity only. Only the Principal Investigator (PI) and Study IG Lead are authorised to move data in or out of the DSH. Any person in addition to the PI or study IG lead requesting access to the PI’s project folder within the DSH will require approval by either the PI or study IG lead, and authorisation by the RGEIT.

No personnel outside the direct employment of the University of Manchester will have access to the NHS Digital data.

NHS Digital data is downloaded into the DSH via a single point - a static IP addressed laptop within the RGEIT office.

For this study, NHS Digital data will be downloaded by the PI. The PI will create a separate password-protected sub-folder within their project folder, which can only be accessed by either the PI or Co-study IG Lead. After logging on to the laptop with their UoM username and password, the PI will access the DSH using the University’s 2-factor authentication Service (Duo) to verify their identity. Data from NHS Digital is transferred into the separate password-protected sub-folder in PIs project folder within the DSH using the Secure Transfer Service. One the data is stored; this will be manually reviewed by the PI to ensure there are no obvious issues with the data. The PI will then create a separate pseudonymised data file, removing all personal identifiers, leaving the SPN as the only record-level identifier. Once this pseudonymised data file is checked and authorised by both the PI and Co-study IG Lead, the original NHS Digital data file will be destroyed.

Identifiable participant data will not leave the DSH.

The pseudonymised NHS Digital data file will be stored within the PI’s project file in the DSH. Here it will be combined with other study data to create a pseudonymised research database for analysis. This data includes:

NHS Digital data for each participant. This will be combined with data collected from:

1) The participant directly using study questionnaires and physical assessments (recorded onto the case report form)

2) Cardiac device data (SPN-coded)

3) Data from medical records (recorded onto case report forms)

The pseudonymised research database will remain within the DSH.

The pseudonymised research database will be used for the majority of all analysis. Data will then be used to analyse temporal trends in physical activity surrounding non-elective hospitalisation episodes. Analysis will be performed using software within the DSH (mainly Excel, SPSS, R, GraphPad).

Data Flow out with the DSH

In order to facilitate the movement of data out with the DSH, an anonymised research database will be created within the DSH. Anonymization will take place in line with ICO guidance. Any data leaving the DSH will be reviewed by either the PI or study IG lead to ensure it is compliant with UoM guidance, HES analysis guidance and ICO anonymization guidance all releases of data will be aggregated with small numbers suppressed in line with the HES analysis guide.

No NHS Digital data will be transferred outside the UK.

Data Linkage

NHS Digital data will be combined with data collected as part of the PATTErn study as stated above and explained in the patient information sheet. It will not be linked with any other databases.

Study Participant Data Anonymity

Once data has been pseudonymised the identifiers will be securely destroyed and there will be no attempt to re-identify individual participants.

Additional Information

One aspect of the study will take place out with the UK; however, this will not involve NHS Digital data. Pseudonymised cardiac device download files require processing by the Medtronic technical team whom are based in Maastricht, the Netherlands. These files will be transferred across and back using an online encrypted transfer service. This data will then be added to the pseudo anonymised research database at the University of Manchester. This process does not involve any NHS Digital 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 ie: employees, agents and contractors of the Data Recipient who may have access to that data).

Dissemination of Results

When processing is complete, outputs will include:

1. Submission to peer review journal. The UoM supports open access publication.

2. Presentations at:

a) Internal meetings at UoM and MFT

b) Seminars and workshops at academic events (where appropriate)

c) National and international academic conferences

d) Patient engagement events

3. A ‘lay summary’ of results for study participants and interested non-academic parties. This is anticipated to take the form of a 2-page pdf document, which will be disseminated via:

a) letters/emails to study participants

b) PATTErn study webpage on publicly accessible UoM website

c) British Heart Foundation literature/website/online publication

4. Reports to:

a) UoM – in form of progress reports (study sponsor), PhD thesis and coursework

b) British Heart Foundation – in form of progress reports (study part-funder and PhD fellowship funder)

c) Medtronic Inc. – in form of progress reports (study part-funder)

The target time for release of these outputs is April 2021.

Level of data contained in outputs

The majority of outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide. Some ‘case study’ timelines for individual participants may be published to visualise the association between physical activity and non-elective hospital admission/attendance events – however all dates will be removed (dates replaced with timeline of day 0-365) and no data will be presented which could directly identify the individual e.g. specific details of reasons for hospital attendance/admission or multiple co-morbidity data. This is considered essential for displaying the clinical significance of activity data.

Dissemination and communication approach

Through the outputs stated above, the results of the PATTErn study will be disseminated to target audience of researchers, data scientists, academic groups, innovative technology-focused organisations and research participants. The British Heart Foundation (BHF) has established links with a wide variety of communities – and will help ensure output is brought to the attention of policy makers.

Communication channels will include

Academic channels – peer review journal manuscripts, presentation at conferences

Website and newsletters – via UoM, BHF and MFT

Open lectures and talks – public promotion of results via the Manchester Institute for Collaborative Research on Ageing (MICRA). We will present the results of the PATTErn study at least one MICRA event.

Exploitation of results/outputs

The UoM has a research collaboration agreement in place with Medtronic outlining data and knowledge ownership and access rights. In line with UoM postgraduate research policy, all newly created knowledge, tools or technologies will be under the ownership of the UoM. Results will be shared with all stakeholders (including Medtronic) as stated above in ‘Level of data contained in outputs’. Funders will have no influence on the study results.

As part of the study inclusion/exclusion criteria, patients must be able to provide written, informed consent in the English language in order to be recruited. This restriction was put in place primarily due to the lack of resources at the recruiting site to (a) provide a service to translate all study documents and (b) provide a face-to-face translator for study consent and assessment. Outputs will be published in the English language only. These factors will restrict participation and dissemination of work to English speaking members of the public.

The impact of the English language only consent on the outputs of the study is that they may well be limited to only reflecting outcomes for a cohort of patients who are able to provide written, informed consent in the English language and the benefits of the study may also be limited to this cohort and may not be able to be extrapolated across to reflect the broader population.

The main benefit of this study will be improved knowledge of the association between physical activity and hospitalization events in older people with cardiac devices. This will be of benefit to health and social care as outlined below:

Researchers: Improved knowledge in this research area opens up new avenues of research within the field of ageing/heart failure research. If physical activity from cardiac devices is shown to correlate with hospitalization events, this may justify future studies investigating the impact of interventions based on this data e.g. physical rehabilitation, referral to Geriatric Medicine services. The PI will submit the results of the study as part of her PhD. Target date: PhD end date, November 2021, future research strategy plan January 2022.

Clinicians: A greater understanding of the significance of physical activity data from cardiac devices will help health professionals make the best use of cardiac device data to serve their patients. For example, this study may lead to a greater emphasis on recording and reviewing activity data during routine device reviews. Further, down the line, this may lead to service improvements projects using activity data as a trigger for signposting to services such as community physiotherapy or exercise groups. Target date: Dissemination of work to improve knowledge of relevant health professionals: October 2021.

Patients: This work will hopefully lead to improved patient care. This may not occur as a direct result of this study, rather further down the line as results inform service improvement projects. Participants in the study whom have consented to receive information on study results will be better informed on the significance of physical activity, and effort will be made to emphasize the importance of physical activity for the maintenance of heath and function. Recruitment aim = 150 patients. Target date for postage of lay summary to participants: April 2021

As per section 5(c) only patients able to consent in the English language are eligible for recruitment into the study, and all research outputs will be published in the English language only. The impact of this is that participation in the study and availability of results will be restricted to English speaking patients/members of the public.

The benefits of the study may not be able to be extrapolated to reflect the broader population as the study has focused on a patient cohort who are able to provide written, informed consent in the English language.

The purpose of this request is to disseminate an additional field (ADMIMETH) to identify elective or non-elective admissions within the HES Admitted Patient Care dataset. This data is required in order to fulfill the purpose outlined in this agreement. At this current stage no Yielded Benefits have been realised due to not receiving the required data for processing.

The Data Controller will update the Yielded Benefits following processing of the required data should a further renewal of the agreement be required.