Prevention of Shoulder Problems Trial: exercise to prevent shoulder problems in patients undergoing breast cancer treatment.

The University of Warwick Clinical Trials Unit will be responsible for processing the data.

The PRevention of Shoulder ProblEms TRial (PROSPER) is a multi-centre pragmatic two-arm Randomised Control Trial (RCT) examining an intervention to prevent shoulder problems in patients following breast cancer surgery. Following breast cancer surgery, it is common for women to suffer a range of postoperative symptoms in the upper arm and shoulder. These postoperative symptoms can persist for many years after treatment. The intervention in PROSPER builds upon existing evidence that exercise programmes following surgery can improve functional outcomes and reduce complications. The purpose of the PROSPER trial is to examine whether such a programme is clinically and cost-effective in comparison to usual care.

The Warwick Clinical Trials Unit has recruited approximately 380 participants into the PROSPER study. For these participants, the researchers hold identifiable data, this includes, name, postcode, date of birth, NHS number and other self-reported data including demographic characteristics, health-related quality of life, and service use. Participants have consented to the researcher holding these data and data processing regulations will be adhered to. Recruitment into the trial has now finished.

The two trial arms that are being tested within this trial are as follows:

Usual care arm: Usual care (control arm) participants continue to receive current care. Current care is characterised by patients receiving information leaflets that recommend movements and exercises that can be conducted to improve outcomes following surgery.

Treatment arm: The treatment arm (intervention arm) participants additionally receive a physiotherapist-led individually tailored exercise programme and a series of appointments to monitor and assess recovery.

The HES data requested from NHS Digital will be used to facilitate and validate the costing of the PROSPER randomized controlled trial. As part of the trial, a comprehensive economic evaluation is being conducted. The goal of the economic evaluation is to estimate the cost-effectiveness of the PROSPER intervention compared to usual care. To achieve this, a systematic comparison of both the costs and outcomes associated with the intervention relative to usual care will be conducted. Thus, there are two key components of interest: costs and outcomes.

Relevant outcomes on health and social care are being collected prospectively within the trial. To calculate the costs associated with the intervention, trial participants are completing resource use questionnaires at baseline (randomisation), and the two follow-up time points (six months and 12 months). The resource use questionnaires seek to identify all the healthcare usage; this includes hospital care, community care, social care, and medicines and drugs. Resource use questionnaires rely on patient recall and therefore may be subject to recall bias whereby individuals fail to accurately identify the resources they used in a given time frame.

The objective of processing the HES data will be to validate the hospital data collected via resource use questionnaires within the PROSPER study.

The University of Warwick require data from three different data-sets (APC, Outpatient, and Critical Care) for three years; 2015/16, 2016/17 and 2017/18 to link to trial data to allow analysis of service use for the trial cohort.

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).

The processing activities are as follows:

1) Patient identifiers are being recorded for participants within the PROSPER trial at The University of Warwick Clinical Trials Unit (WCTU).

2) The University of Warwick will send in the following identifiers to NHS Digital for linkage to the HES Data:

• NHS number

• Date of birth

• Postcode

• Unique Study ID

3) NHS Digital will be responsible for linking the patient identifier(s) to HES data. NHS Digital will return pseudonymised HES data, with the unique study ID linked to the PROSPER cohort. This will allow the researchers to validate responses from the questionnaires.

4) Warwick Clinical Trials Unit will store these data on a server at the University of Warwick which can only be accessed by those granted authorisation by the Chief Investigator. All staff involved in the study at the Warwick Clinical Trials Unit are substantive employees at the University of Warwick.

5) The University of Warwick health economists will calculate costs of hospital care using the HES data for participants within the PROSPER trial. The HES data costs will be used to assess the validity of the PROSPER resource use data.

There will be no requirement or attempt to re-identify individuals from the data. HES data will be stored separately from any data that allows identification, other than Unique Study ID which will be used to link the questionnaire data.

The University of Warwick require data from three different data-sets (APC, Outpatient, and Critical Care) for three years; 2015/16, 2016/17 and 2017/18 to link to trial data to allow analysis of service use for the trial cohort.

The first dissemination of data will be just for the years 2015/2016 and 2016/2017. This will provide baseline and six month follow up data for approximately 100 patients.

The second download will include 2017/2018 data. This will provide baseline and six month data for all participants, as well as 12 month follow up data for approximately two-thirds of the patients within the PROSPER trial.

Data for 2018/19 which would give the final time-point for the other third of participants are not being requested as it would not allow time for analysis before the study ends. The University of Warwick have been selective in the variables requested; choosing only those which may aid the analysis. No unnecessary sensitive variables are being requested. Any potentially sensitive data e.g. data relating to Trusts will not be published.

The University of Warwick aim to examine whether there is any systematic bias within the self-report data. Trial data will rely on self-report resource usage, however, this is subject to recall bias. The University of Warwick will therefore use HES data alongside trial data to calibrate trial costs, there will be no linkage of any other trial data to the HES data.

Costing using the HES data will require the use of the reference cost grouper. The costs as calculated through trial data will be compared to that of the HES data. The University of Warwick will assess whether there is concordance between the HES data and the self-report data. If any systematic discordance is found between the two data sets then resource use for those who the researchers do not have HES data for will be adjusted accordingly.

All outputs will be restricted to aggregated data with small numbers suppressed in line with the HES Analysis Guide.

The data from NHS Digital will not be used for any other purpose other than that outlined in this Agreement.

The data from NHS Digital will not be linked to any other data other than those outlined in this Agreement.

Warwick Clinical Trials Unit will store these data on a server at the University of Warwick which can only be accessed by a restricted number of personnel.

This is an NIHR funded study and all outputs of the trial will therefore be reported to the NIHR. The NIHR will publish the report of the PROSPER trial, including the economic evaluation, on its website.

Specific outputs include:

- The final NIHR Health Technology Assessment (HTA) Mongoraph; which will be published in 2019.

- Peer reviewed publications relating to the economic evaluation of the PROSPER trial. This will be alongside the primary clinical findings, or as a standalone paper. These will be submitted at the end of 2018 and published in 2019.

- Main trial findings will be summarised on the Warwick Clinical Trials Unit main website.

- Findings will also be disseminated via social media, this includes the Warwick Clinical Trials Twitter

page (@WarwickCTU) and Warwick Health Economics Twitter page (@HealthEconWMS).

A methodological paper comparing costing methodologies will also be considered.

The study findings will be widely promoted on Twitter and usual scientific routes, thus via national and international academic conferences and peer-reviewed publications. There are patient representatives within the research team who will also be involved with preparing materials for wider dissemination. Findings will feedback to patient groups who provided input to the early development phases of the trial.

The primary benefit is to fulfil a key requirement of the PROSPER trial, namely to estimate the cost-effectiveness of the physiotherapy intervention, compared to routine care. NICE can use this information when deciding whether to recommend physiotherapy after breast cancer surgery.

Given the frequent complications following breast cancer surgery, there is great potential for the intervention to save the NHS money in the long run, whilst improving outcomes for patients. To ensure the limited budget of the NHS is being spent efficiently, it is important to conduct a thorough examination of the cost-effectiveness of the intervention. The HES data requested should help inform an accurate assessment of the cost-effectiveness of the PROSPER intervention, and ultimately inform changes in clinical practice in the UK.

The HES dissemination of follow-up data for trial participants will be obtained during 2018 and the University of Warwick will aim to have final results and outputs ready for wider publication during 2019. The report to be published by the National Institute of Health Research HTA Programme, and linked peer-reviewed journal articles, will provide critical knowledge on the efficient delivery of shoulder rehabilitation following breast cancer surgery.

Shoulder pain is extremely common following breast cancer surgery. Should the intervention be successful there is potential to achieve great benefits in terms of reduced pain and increased mobility following breast cancer surgery. The researchers will be in a position to promote the findings of the study but unfortunately do not have the capacity or the funding to ensure the implementation beyond the lifetime of the clinical trial.

Yielded Benefits is not a requirement for new applications.