NHS Digital Data Release Register - reformatted

Queen Mary University of London

🚩 Queen Mary University of London received multiple files from the same dataset, in the same month, both with optouts respected and with optouts ignored. Queen Mary University of London may not have compared the two datasets, but the identifiers are consistent between datasets for the same recipient, and NHS Digital does not know what their recipients actually do.

Project 1 — DARS-NIC-55950-Y5Y2Y

Opt outs honoured: Y

Sensitive: Non Sensitive, and Sensitive

When: 2017/06 — 2017/08.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012, Section 42(4) of the Statistics and Registration Service Act (2007) as amended by section 287 of the Health and Social Care Act (2012)

Categories: Anonymised - ICO code compliant, Identifiable

Datasets:

  • Hospital Episode Statistics Admitted Patient Care
  • Office for National Statistics Mortality Data

Objectives:

More than 150,000 high-risk patients undergo emergency surgery each year in the NHS, following which at least 90,000 patients develop complications resulting in over 20,000 deaths before hospital discharge. Patients who develop complications but survive, require in-hospital care for prolonged periods, suffering substantial reductions in functional independence and long-term survival. Recent data show that abdominal surgery and the need for surgery on an emergency basis are amongst the strongest factors associated with poor post-operative outcome. Around 35,000 patients present to NHS hospitals each year with precisely this pattern of risk and undergo a procedure known as ‘emergency laparotomy’. This term describes a major surgical procedure to treat an acute and often life threatening problem with the gut or other abdominal organ. Around 180 patients undergo emergency laparotomy in a typical NHS hospital each year with a 90-day mortality of 25%. There is considerable heterogeneity in standards of care between hospitals, including wide variations in the involvement of senior surgeons and anaesthetists and post-operative admission to critical care, which are associated with important differences in mortality rates. In 2010 the Department of Health commissioned a Royal College of Surgeons of England (RCS) working group to develop an integrated care pathway which could improve the quality of care for patients undergoing emergency laparotomy. An integrated care pathway was defined which represented an optimal standard of peri-operative care deliverable in all NHS hospitals. Examples of interventions included consultant led decision making and treatment, standards for diagnostic testing, structured post-operative surveillance, time limits for review of deteriorating patients and early admission to critical care. To date, there has been little systematic implementation of any component of the integrated care pathway. Most opinion leaders agree there is an urgent need for a national project to improve survival for emergency laparotomy patients. However, there is uncertainty about how best to achieve such improvement. Some question the benefits of quality improvement initiatives, pointing to the lack of robust clinical evidence of effectiveness, both in terms of generic methodologies advocated to improve quality, and the specific changes in patient care. There are examples where a discrete quality improvement intervention was associated with improved clinical outcomes. And experience from recent quality improvement initiatives has shown that challenges with quality improvement programmes can be overcome. However, doubts over the clinical effectiveness of quality improvement projects continue to limit the success of these initiatives. There is a clear need for robust clinical evidence to support or refute the use of this approach to improve clinical practice and, ultimately, patient outcome. The Healthcare Quality Improvement Partnership (HQIP) has commissioned a new National Emergency Laparotomy Audit (NELA), providing a unique opportunity to study the clinical effectiveness of a quality improvement project to implement an integrated peri-operative care pathway for emergency laparotomy patients. By providing a robust evidence base for quality improvement in peri-operative care, the findings of this work could accelerate implementation of care pathways for all categories of high-risk surgery with the potential for widespread improvements in survival affecting more than 170,000 NHS patients each year. QMUL have conducted a large pragmatic clinical trial of the effectiveness of a quality improvement project to implement a modified version of the RCS integrated care pathway to improve patient outcomes following emergency laparotomy. The trial is called Enhanced Peri-Operative Care for High-risk patients (EPOCH). Our aim is to provide the definitive evidence needed to inform practice in this area. TRIAL OBJECTIVES: 1. To evaluate the effect of a quality improvement intervention to promote the implementation of an integrated peri-operative care pathway on survival at 90 days following emergency laparotomy. 2. To assess the cost-effectiveness of the quality improvement intervention compared to ongoing clinical practice without the intervention. 3. To evaluate the long-term effects of the intervention on standards of care and mortality following emergency laparotomy in participating hospitals. Primary outcome measures: All cause mortality at 90 days following surgery Secondary outcome measures: All cause mortality at 180 days following surgery, duration of hospital stay and hospital re-admission within 180 days of surgery. In eight hospitals EQ-5D 3L will be collected and healthcare resource use data preoperatively, and at 90 and 180 days after surgery to perform a health economics analysis. In eight hospitals a Health Economics sub-study has been conducted. This involved patients who are already part of the NELA audit. QMUL consented patients and collected information (at 30, 90 and 180 days after surgery) about their quality of life and healthcare use using standardised questionnaires. The information that was recorded on paper questionnaires was entered into a secure, password protected database hosted by Queen Mary University London. No identifiable information was recorded in the database. Each patient in the Health Economics sub-study has a NELA ID which will be used to link with the main NELA dataset and data from ONS and HES. The team will collect EQ-5D 3L (a standardized questionnaire for measuring generic health status and quality of life) and healthcare resource use data preoperatively, and at 90 and 180 days after surgery to perform a health economics analysis which aims to assess whether implementing the quality improvement intervention is likely to be cost-effective on average and whether this varies between low and high mortality groups. As a national clinical audit funded by the Healthcare Quality Improvement Partnership (HQIP), the National Emergency Laparotomy Audit (NELA) has been established with the fundamental aim of improving the quality of patient care, and clinical outcomes. The EPOCH trial is an important example of how the data collected by NELA can be used to drive improvements in patient care.

Expected Benefits:

NELA has become a successful audit and is already tracking some improvements in patient outcomes nationally. However feedback from hospitals is that it can be difficult for them to know what changes to make that are most likely to lead to sustained improved outcomes for patients, or how best to bring about these improvements. EPOCH is built around a trial of a systematic and specific improvement intervention: a quality improvement programme. We have undertaken and will publish a great deal of evaluation of the quality improvement programme itself (both the ethnography and also through a process evaluation of the trial). Access to HES and ONS data is essential so that we can analyse and report the impact of the improvement intervention in terms of quality of care and patient outcome. Once we know the outcomes of the trial, we will be able to deduce evidence-based recommendations about what works and what does not work in improving care for this patient group, thereby allowing frontline clinicians and managers to make informed decisions about how best to improve patient care. Further to this, we will gain an understanding of quality improvement as a process in the NHS (in terms of feasibility and impact) which will be informative for commissioners and policymakers. Due to the synergy between EPOCH and NELA, with the latter providing an ongoing data set to support quality improvement, we could expect to begin to see this accelerated improvement commence within months of publication of the trial results. This is because hospitals will be able to track their improvements through ongoing participation in NELA. Publications will allow widespread dissemination of the findings amongst health professionals. Publishing in peer-reviewed journals will allow greater discussion of the strengths and weaknesses of the results, and will provide the benefit of peer-review of the work from third parties. Through dissemination in publications and at meetings (please see ‘Outputs’) results about did and didn’t work in the clinical care pathway will reach frontline clinicians (surgeons, anaesthetists and intensive care doctors). Therefore, we would expect to see measurable benefits, for emergency laparotomy patients nationwide, within 6-12 months of publication. We believe this work will influence trigger a sustained and widespread peri-operative quality improvement for all high-risk patients over a much longer period, with sustained and measurable benefits seen over a 2-10 year period from publication. A further measurable benefit is the anticipated contribution of the outputs produced as a result of the trial to clinical guidance and national policy within a 1-2 year period from publication.

Outputs:

The following outputs are planned: Final Report to Funder - 30th May 2017 Final EPOCH Investigators Meeting - summer 2017 Peer Reviewed Publications - summer 2017 Conferences and meetings - 2017 - ongoing Media – 2017 – ongoing NELA Website - 2017 Details of expected outputs: The main scientific report containing only aggregated data with small numbers supressed in line with the HES analysis guide will be sent to the Lancet journal for their consideration first. The Lancet have already published the trial protocol, and have expressed an interest in the work. Second choices would include the British Medical Journal, the New England Journal of Medicine, and the Journal of the American Medical Association. These are all general interest journals read by a wide range of healthcare workers worldwide. QMUL plan to invite all the co-investigators (from 92 hospitals) to a final EPOCH meeting where we will disseminate and discuss the findings. QMUL will also present the work at scientific meetings, and congresses. For example the ‘Annual Congress of Enhanced Recovery and Perioperative Medicine’ and the ‘Peri-Operative Medicine Congress’. QMUL will also disseminate the aggregated findings through the mainstream media (e.g. BBC news, Channel 4 News, Times Newspaper etc.) and also through social media (e.g. Twitter) with the support of our patient representatives. The main target audience will be surgeons, anaesthetists and intensive care doctors but also patients and their carers. The findings will also be published on the NELA and EPOCH websites. All outputs will aggregated with small numbers supressed in line with the HES analysis guide.

Processing:

The processing will follow these steps: Step 1 - RCS flow NHS number to NHS Digital, NHS Digital flow HES + ONS data to RCS with Objections upheld, this step has been completed under the agreement NIC-355855 and the flow is covered by s251 of the Care Act 2006. Step 2 - QMUL apply to NHS Digital for HES + ONS data (this application). The legal basis for the HES APC data is The Health and Social Care Act 2012 and the legal basis for the ONS data is s42(4) of the SRSA 2007. Step 3 - NHS Digital confirm to HQIP that QMUL have approval for the release HES + ONS data, once the application has been recommended for approval by DAAG Step 4 - RCS flow HES ID + study ID to NHS Digital to define cohort Step 5 - NHS Digital send study ID’s to RCS with objections upheld if relevant (copy in HQIP) Step 6 - QMUL apply separately to HQIP for audit data Step 7 - NHS Digital flow HES + ONS data to QMUL, (with study ID) with objections upheld if relevant Step 8 - HQIP authorise RCS to flow the audit data to QMUL Step 9 - RCS flow authorised audit data to QMUL Step 10 - QMUL will perform linkage of HES/ONS and Audit data based on a study id Step 11 - QMUL will perform linkage of HES/ONS and Audit data with Heath Economic sub-study data based on a study id The only identifiable field received by the EPOCH team is ONS Date of Death. The full Date of Death is required to be able to calculate survival at 2 time points (90 day, 180 day). The EPOCH team will not use ONS Date of Death to identify any individual patients. To be clear no attempt will be made at any point to re-identify and individual. In line with good practice, QMUL retains raw data to enable responses to journal peer reviews and re-analysis if required. Therefore Date of Death and Cause of Death will be retained for a 2 year period before being destroyed. In line with QMUL procedure, all other data will be archived for 20 years. The data received from HSCIC will not be linked back to the identifiable NELA database. An extract of annonymised NELA data will be linked to the HES-ONS data via the NELA ID. The annoymised NELA data and the HES-ONS data will also be linked to a Health Economics (HE) dataset via the NELA ID. The HE dataset does not contain any identifiers. Each patient in the Health Economics (HE) dataset also exists in the NELA dataset, and their NELA ID has been recorded in both datasets. The data has been collected as part of an ethically approved sub-study for the EPOCH trial, using patient consent. The EPOCH team Statisticians and Health Economists who will work on the linked dataset do not have access to the identifiable data set held by the RoCA or any identifiers held locally at hospitals. Only Queen Mary University Staff will have access to the data. The EPOCH team at QMUL will analyse the data in a secure environment with restricted access (called a safe haven) to assess the objectives listed. Outputs from the EPOCH analysis will only include aggregated data, no individual level data will ever leave the safe haven where the analysis is taking place .


Project 2 — DARS-NIC-37666-Z1C3M

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

Sensitive: Sensitive, and Non Sensitive

When: 2019/02 — 2020/10.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(7), National Health Service Act 2006 - s251 - 'Control of patient information'.

Categories: Identifiable

Datasets:

  • MRIS - Flagging Current Status Report
  • MRIS - Cause of Death Report
  • Hospital Episode Statistics Outpatients
  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Admitted Patient Care
  • MRIS - Cohort Event Notification Report
  • Demographics
  • Civil Registration - Deaths
  • Cancer Registration Data

Objectives:

Breast cancer is increasingly becoming a “survivable” disease, and an increasing number of recurrences occur late, so that there is much interest about the long-term efficacy and safety of treatments. The Oxford Overview process has provided useful data on follow-up for 15-20 years after tamoxifen therapy. Such data are lacking for the newer aromatase inhibitors (AIs). This study provides a unique opportunity to address this issue. The ‘Arimidex, Tamoxifen, Alone or in Combination’ (ATAC) trial is the vanguard breast cancer trial for the use of AIs in the adjuvant setting, with a median follow-up of 100 months. Established in 2010, the Long-term Anastrozole vs Tamoxifen Treatment Effects (LATTE) observational study aims to collect vital long-term safety and efficacy data on anastrozole and tamoxifen, drugs that were taken by post-menopausal women as part of the Anastrozole Tamoxifen Alone or in Combination (ATAC) trial. ATAC was a crucial breast cancer trial evaluating the use of aromatase inhibitors (AIs) in the adjuvant setting, with a median follow-up of over 10 years. The ATAC trial assessed the safety and efficacy of anastrozole+placebo, tamoxifen+placebo and anastrozole+tamoxifen in post-menopausal women undergoing treatment for invasive primary breast cancer The LATTE study aims to collect further follow-up information on a maximum of 2200 UK eligible patients, who were randomised to the monotherapy arms (anastrozole or tamoxifen) in the ATAC trial. The study’s primary objective is to provide additional efficacy and safety data on time to recurrence of breast cancer and death after recurrence. The secondary objectives include time to distant recurrence, cancer-specific survival, new breast primaries, other cancers, ischaemic cardiac and cerebrovascular events and hip (and other) fractures. Participants will be followed-up annually until at least 15 years median follow-up. It should be noted that the Research Ethics Committee (REC) and Confidentiality Advisory Group (CAG) have already approved the LATTE study for collection of vital HES, cancer registration and mortality data sets as part of this study. In 2010, results from 10-year follow-up from ATAC demonstrated that there is long-term superiority of anastrozole over tamoxifen as initial adjuvant therapy for post-menopausal women with hormone-sensitive early breast cancer. This paper demonstrated that there were significantly lower rates of local and distant recurrence, as well as reduced contralateral breast cancer in patients treated with anastrozole. However, in 2016 a recent preliminary analysis of the LATTE study demonstrated that the benefits of anastrozole do not continue long-term after 10 years in terms of reduced recurrence rates. The results also suggested that there was a larger reduction of new contralateral tumours with tamoxifen. However, the preliminary analysis was limited in that there was limited data with possible under reporting of vital events inked to the safety and efficacy endpoints; this therefore may diminish the effect of anastrozole. Queen Mary University of London (QMUL) therefore require extended cohort events, obtainable from HES, cancer registration and mortality data in order to conclude the long-term safety of either intervention. QMUL will then be able to integrate the pseudonymised LATTE data with the ATAC data sets in order to perform an analysis of the overall 20-year follow-up period.

Yielded Benefits:

Registry data provided to date has identified numerous cases of breast cancers, other cancers and the effects of long term treatment in participants within the LATTE cohort, with whom the Centre for Cancer Prevention at Queen Mary University of London have no direct contact and therefore are not already known by the study team. Collection of this data has contributed directly to the primary endpoints of the trial, providing us with information on mortality and recurrence of breast cancer. The ATAC trial which is the vanguard breast cancer trial for the use of AI's in the adjuvant setting, with a median follow up of 10 years. The LATTE study proposes to collect further follow up information on as many eligible participants as possible who were randomised to monotherapy (anastrozole or tamoxifen). The results of the main analysis of the ATAC trial have formed the basis for the decision by the National Institute for Health and Care Excellence (NICE) to include anastrozole in their guidelines, recommending anastrozole as a first line treatment for post-menopausal women with early stage breast cancer. The continuing follow-up in the LATTE study will enhance our understanding on long term effects of anastrozole and tamoxifen in terms of secondary prevention and late adverse events. These data will help to further understand who benefits the most from taking these drugs and hence will form evidence for clinical/oncological decision making.

Expected Benefits:

The data sets returned by NHS Digital will provide vital information on side effects, survival and breast cancer incidence. LATTE is currently the only study assessing the long-term effects of an aromatase inhibitor (AI) compared to tamoxifen, achieving a 20-year median follow-up period. Using the requested data sets, the LATTE Central Coordinating Office (CCO) will conduct research and analyses that will help provide further information on: - the efficacy of tamoxifen in reduction of breast cancer recurrences in post-menopausal women initially diagnosed with invasive breast cancer; - the efficacy of anastrozole in reduction of breast cancer recurrences in post-menopausal women initially diagnosed with invasive breast cancer; - the efficacy of anastrozole compared to tamoxifen in reduction of breast cancer recurrences in post-menopausal women initially diagnosed with invasive breast cancer; - the effect tamoxifen and anastrozole on all-cause mortality rate, cancer incidence, distant recurrence and new breast primaries; - the safety of tamoxifen in terms of occurrence of ischaemic cardiac and cerebrovascular events, as well as hip (and other) fractures; - the safety of anastrozole in terms of occurrence of ischaemic cardiac and cerebrovascular events, as well as hip (and other) fractures; - the safety of anastrozole compared to tamoxifen in terms of occurrence of ischaemic cardiac and cerebrovascular events, as well as hip (and other) fractures. The research will help inform public health bodies such as NICE on the long-term safety and efficacy of tamoxifen and anastrozole as adjuvant therapy. These analyses will enable patients and women who have been diagnosed with invasive breast cancer to make a more informed decision on the adjuvant treatment options available to them. The target data for these benefits is expected around 2020.

Outputs:

Data from a preliminary analysis was presented at the San Antonio Breast Cancer Symposium in December 2016 as a poster. QMUL plan to publish the main results from the LATTE cohort within 12-18 months after receipt of the NHS Digital data. This will enable them to report a more complete data set, with up-to-date information on recurrences, deaths, other primary cancers, and adverse events. To demonstrate a potential carry-over effect of anastrozole versus tamoxifen, long-term follow-up in excess of 10 years is essential. Therefore, after the initial publication in approximately 2020 QMUL plan to publish an updated manuscript on the benefits of AI therapy in 2022/2023. The main target audience for their data outputs are the patients with early breast cancer, and the medical professionals who can use the results for decision making. Further outputs and analyses of the LATTE study will be published in peer-reviewed medical journals such as The Lancet, The Lancet Oncology, Breast Cancer Research and Treatment and the British Journal of Cancer. Outputs are expected in 2020 when there is 10 years' of follow-up data available. The public will be made aware of the research progress through the following web page: https://www.qmul.ac.uk/wolfson/research-projects-a-z/current-projects/ The anticipated audiences of the above outputs are researchers, scientists, stakeholders for the wider project and research participants. Policy makers and public health bodies such as NICE will be targeted with the outputs of the study. NICE will be advised on the long-term safety and efficacy of tamoxifen and anastrozole as adjuvant therapy. These analyses will enable patients and women who have been diagnosed with invasive breast cancer to make a more informed decision on the adjuvant treatment options available to them. Data contained in any outputs will be aggregated with small numbers suppressed in line with the HES analysis guide. The funder will not have influence on the outcomes nor suppress any of the findings of the research.

Processing:

All organisations party to this agreement must comply with the Data Sharing Framework Contract, including requirements on 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). The Centre for Cancer Prevention at QMUL will send LATTE cohort data, i.e. a list of LATTE study participants, to NHS Digital for linkage. The following identifiers will be sent to NHS Digital by the Centre for Cancer Prevention (CCP) team at QMUL: -Full Name, -Date Of Birth, -NHS Number, -Postcode, -Study ID. The data received by Centre for Cancer Prevention at QMUL will not be used for any purpose other than to meet objectives as stated in the trial protocol and will not be shared with any other third party or organisation. The data sets returned by NHS Digital to QMUL will contain identifiable data. This is so that QMUL can confirm the accuracy and strength of each linkage for LATTE participant in the study cohort. This ensures that all side effects, survival and cancer incidence data are correctly attributed to each LATTE participant, thus avoiding any errors which would invalidate the study. On receipt of the returned data set from NHS Digital, the Centre for Cancer Prevention at QMUL will link this data set to the data set contained in the applicant's LATTE study database via the full name, date of birth, NHS number and study ID. As mentioned, the data linkage enables the correct identification of the LATTE patients from the data set sent by NHS Digital. Once the patients are identified, researchers at QMUL will then update the participant entries in the LATTE study database with date and cause of death, cancer recurrence, diagnosis and coding, and any HES data relevant to the study (fractures, cardiovascular and thrombo-embolic events). Following the linkage and participant identification, all identifiers except the study ID will be removed from the data set returned by NHS Digital and stored as a reference for analysis until study completion. Processed data sets with all identifiers removed will be retained for analysis until study completion, following which it will be archived for 20 years. The original 'raw' data set returned by NHS Digital containing identifiable data will be stored on a separate server meeting NHS Digital security requirements, as per DARS Guidance Notes on Security for where Mortality Data is involved. Personal identifiable data are stored on the secure network and never on the local drives of unencrypted QMUL desktop PCs. QMUL creates back-ups of the data stored on its network. The encrypted backup tapes are stored offsite with Iron Mountain UK Ltd. The data is not accessible by any Iron Mountain employee. All personal identifiable data received at the Centre for Cancer Prevention are stored electronically on a database, local to and administered by substantive employees of QMUL only (no third parties). Personal identifiers of study participants are stored separately to the clinical data, with access strictly restricted to only QMUL substantive members of staff. Only these members of staff will have access to the personal identifiable data. Additionally, access to the personal identifiers on the separate database is controlled by separate username and password access and is controlled by the IT Department. The separate Mortality and HES servers are self-contained within the Centre for Cancer Prevention network within the QMUL network. They are firewalled from external connections and the rest of the network. All traffic through the checkpoint firewall is logged. Access to data will be from within the network using workstations that have a currently supported operating system which includes security patches. Users are not permitted to download the data from these workstations. No remote access (i.e. through a VPN/RDP connection) is permitted. No mobile devices will be used to access any data. The network is externally scanned on a regular basis. The 'raw' data sets will be stored until the data destruction date as per the Agreement with NHS Digital. The storage architecture is compliant with the NHS DIGITAL data sharing framework contract and once QMUL is issued with a Data Destruction notice, the data from all storage including backups can be securely and permanently removed within 14 days. Data can be securely wiped to NHS Digital standards (multi pass pattern wiped to at least HMG S5 Enhanced on site and if end of life, degaussed and physically destroyed). Under this agreement, it is not permitted to share the "raw" data with third parties. Only aggregated data with small numbers suppressed in line with the HES Analysis Guide is permitted to be shared.


Project 3 — DARS-NIC-324220-P6W9Y

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

Sensitive: Sensitive, and Non Sensitive

When: 2016/04 (or before) — 2019/08.

Repeats: Ongoing, One-Off

Legal basis: Section 251 approval is in place for the flow of identifiable data, Approved researcher accreditation under section 39(4)(i) and 39(5) of the Statistical Registration Service Act 2007 , Health and Social Care Act 2012 – s261(7)

Categories: Identifiable, Anonymised - ICO code compliant

Datasets:

  • MRIS - Flagging Current Status Report
  • MRIS - Cause of Death Report
  • Hospital Episode Statistics Admitted Patient Care
  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Outpatients
  • MRIS - Cohort Event Notification Report
  • MRIS - Scottish NHS / Registration
  • MRIS - Members and Postings Report

Objectives:

Purpose The Data Recipient agrees to process the Data only for the following purposes agreed with the HSCIC: Objective for processing: The International Breast Cancer Intervention Study (IBIS-II) Prevention & DCIS (Ductal Carcinoma In-Situ) studies were designed to continue the work started by IBIS-I in determining whether a chemo preventive strategy towards breast cancer is beneficial. Both are double-blind, placebo controlled, randomised trials, which recruited post-menopausal women between the ages of 40 & 70. It is the intention to collect data for these cohorts regarding breast cancers, including recurrences, through the use of HES and ONS data, in order to assess efficacy, and meet the primary and secondary objectives of the trial. The primary objective is to determine the long-term effect of using anastrozole and assess if it is at least as effective as tamoxifen in local control and prevention of contra-lateral disease in women with locally excised estrogen receptor (ER) or progesterone receptor (PgR) positive cancer tumors. The participants HES/ONS data will be used to compare the effects and effectiveness of the two drugs used in the intervention (tamoxifen and anastrozole). The study will observe the patients data for any cancer occurrences or serious medical illnesses that occurred each year during the study (intervention arm 2000-2010) tracking notable changes in health their status. Secondary objectives of the study are: i. To examine the rate of breast cancer re-occurrence and new contra-lateral tumors after cessation of tamoxifen or anastrozole. ii. To examine the effect of tamoxifen versus anastrozole on breast cancer mortality Data received from the HSCIC will be used to examine the health statuses of each participant following the end of their taking of the drug to present day and ongoing for long-term analysis. Furthermore, long-term use of anastrozole has been linked to increased fracture rates and research has indicated there are risks of cardiovascular events associated with radiotherapy, a treatment which some of the (Ductal carcinoma in situ) DCIS participants may have undergone. This highlights a vital need for passive follow-up to obtain important safety data regarding other cancers, cardiovascular & thromboembolic events, and fractures, from inpatient, outpatient and A&E records. To measure this the applicant will observe the number of fracture, cardiovascular and thrombolic events that occurred throughout the duration of the study period and after in order to build a profile of possible side effects for each drug.

Yielded Benefits:

Registry data provided to date has identified multiple cases of breast cancers and DCIS events in participants that were lost to follow up and therefore not already known by the trial team. Collection of this data has contributed directly to the primary endpoints of the trial and will continue to provide valuable information on the long term risk/benefit status of the trial interventions. This long term data is essential for informing bodies such as NICE to inform the prescribing utility of these drugs as chemopreventive agents. QMUL have also used mortality data and the targeted code searching of HES to contribute directly to secondary endpoint data. QMUL use registry reported events as a prompt for further investigation by asking participating IBIS-II sites or the participant’s current GP practice to confirm the event and also provide any further information e.g. cancer grade, size, receptor status that is required to meet wider secondary and exploratory objectives. Furthermore, it is anticipated that over the next 18 months that active site follow up could be replaced by registry follow up only. In 2017, an IBIS-II participating site that would have otherwise had to close due to lack of staff resource to deliver the trial was piloted as a ‘registry only’ follow up site. The preliminary data from years 2003-2015 has shown that a higher degree of correlation between the clinical events of interest reported via active follow up methods (questionnaires) and those reported via the registry data. Therefore, active follow up methods (annual questionnaires) have been concluded at the site and instead data collection using registry only data plus verification from the participant’s GP is being used. QMUL hope that this may be rolled out across all sites in England and Wales over the next 18 months so that QMUL can continue to collect essential primary and secondary endpoint data in the most appropriate way given the resource available at participating NHS sites (the trial has now been active for 14 years) and the burden of active follow up procedures on the ageing participant group.

Expected Benefits:

Expected measurable benefits to health and/or social care including target date: 1.1. Primary 1. Following completion of the analysis (duration one year from data received) the customer will be able to publicise the effectiveness of cancer prevention drugs tamoxifen and anastrozole in reducing the occurrence of cancer. 2. To compare any potential or serious side effect from using either tamoxifen or anastrozole for a long period of time 1.2. Secondary 1. To examine the rate of breast cancer recurrence and new contralateral tumors after cessation of tamoxifen or anastrozole. So it can be observed if cancer rates will increase after one has stopped using the medication 2. As the drugs administered in this trial are still widely used it will benefit the public to have a clear understanding if there are any long term issues linked to using the drug and cardiovascular disease, fracture rates and non-cancer deaths. There is considerable interest in the trial from Cancer Research UK who are funding the study the steering committee is co-Chaired by a representative from CRUK. The publication and findings from HSCIC data will be used to inform our studies drug providers the overall impact and effectiveness of their drugs including long-term side effects.

Outputs:

Specific outputs expected, including target date: Data outputs produced will not contain any patient identifiable data or be shared with any third party organisation. The only outputs produced will be publications (medical journals; The Lancet, Breast Cancer research and treatment, British Journal of Cancer) for research purposes using aggregated data or small numbers suppressed derived from our findings at the most. These will be delivered over a 10-year period. The data will not be used for any purpose other than to meet objectives as stated in the trial protocol & will not be shared with any other organisation. Any information which is used for publication in peer reviewed journals will be anonymised & not presented at the individual level .The study estimate that the research will be complete following 9 months after the date that the first dataset received. Any comparisons between findings from other studies or countries will be made the aggregated level and under no circumstances will identifiable data be used for this or in any publications. Only the members of the study will have access to the direct outputs received suppressed in line with HES analysis guide from the HSCIC

Processing:

Processing activities: QMUL will send the IBIS-II cohort data (list of participants on the IBIS-II study) the legal basis for the cohort is Section 251 (Full name, DOB, NHS number) to HSCIC for data linkage. QMUL will use the returned data set to link baseline study findings with potential cancer progression and any occurrence of serious adverse events identified by HES, ONS or cancer registry data. The identifiable data will be used to confirm the strength of each linkage. This ensures that no incorrect diagnoses are made which would invalidate the study. Patient HES/ONS data will be electronically stored on an Oracle database (Oracle 11g). the system is self-contained within the Barts Cancer Centre QMUL network and firewalled off from the rest of the network and external connections. This only allows access for those members mentioned on this applications. No data held from IBIS-I will be used as any part of this study. IBIS-I and IBIS-II are two separate studies run in parallel. QMUL will use the returned data set to link baseline study findings with potential cancer progression and any occurrence of serious adverse events identified by HES, ONS or cancer registry data. Data will be analysed by the study to ascertain how many cancers and/or serious adverse events have occurred since baseline. The applicant will also look at any adverse outcomes such as death in patients who are using anastrozole. This analysis will only make use of the unique study identifier to match each participant to their health record from this stage onwards PID will not be analysed.


Project 4 — DARS-NIC-291938-R6V3V

Opt outs honoured: Y, No - consent provided by participants of research study (Consent (Reasonable Expectation))

Sensitive: Sensitive, and Non Sensitive

When: 2016/04 (or before) — 2021/03.

Repeats: Ongoing, One-Off

Legal basis: Informed Patient consent to permit the receipt, processing and release of data by the HSCIC, Health and Social Care Act 2012 – s261(2)(c)

Categories: Identifiable

Datasets:

  • MRIS - Cause of Death Report
  • MRIS - Cohort Event Notification Report
  • MRIS - Flagging Current Status Report
  • MRIS - Members and Postings Report
  • Demographics

Objectives:

The aim of this study is to establish an epidemiologically based sample of all patients with diabetes within the regions included in the Diabetes Research Network (DRN). By collecting the same non-invasive samples as per clinical care it will be possible to perform tests to look for markers of the complications of diabetes. DNA will be taken from all consenting patients and molecular and genetic information will be combined with clinical information to provide a resource to look for gene/environment interaction in the development of Type 1, Type 2 and other forms of diabetes and their associated complications. Data access is restricted to those named in section 7 of this agreement. Any changes will be notified to the HSCIC.

Yielded Benefits:

The DARE study has contributed to the recruitment of over 25 ethically approved NIHR portfolio studies to date and has involved the participation of over 2,200 people registered on the database. These studies are aimed at either finding a cure for diabetes or improving health of those with diabetes. Having this database has accelerated the research process and provided evidence which has improved health care provision. For example, the team recently searched for a study called REVITA-2. This study for over weight people with type 2 diabetes (T2D) is trialling a new investigative procedure in the small intestine (duodenum). The study is designed to safely alter the inner surface of the duodenum to improve control of blood sugar. The study was very difficult to recruit to. Using DARE the Team were able to identify nearly 200 people meeting the criteria with 5% of these going on to be screened into the study. The benefit for DARE members is that they have access to some of the latest research studies which they may be interested in. The yielded benefit to the company is that they will now be undertaking another study - REVITA -3 which will improve the health of many others with T2D. In another completely different study DARE members were also contacted about a questionnaire - the Diabetes Essential Care Study - which asked their opinion about Diabetes Care provision in primary care. So far nearly 2000 people have responded and presentations have been made on preliminary findings to the study steering committee. The next yielded benefit is to make reports to local CCGs on the findings which could have a beneficial affect on service provision. All outputs will be aggregated with small numbers suppressed in line with the HES Analysis Guide.

Expected Benefits:

The expected benefit of the DARE database is that study team can continue to offer people registered the opportunity to take part in diabetes research. Recruitment is sometimes a difficult aspect of any research and can be costly in terms of searching and advertising. Studies using the database can expect to have an increased chance of recruiting more efficiently and quickly. This speeds up the evidence gained from the research and may lead to prompt changes in treatment and care, improving self management etc. The indirect expected benefit of being on the DARE database is that many members are likely to attend outreach events organised by the study team and have access to new information about their diabetes which in turn improves health and well being for people with diabetes.

Outputs:

The main output is an up-to-date database for the DARE study which means that the study team can contact their members about research opportunities when they become available. They will also be holding a series of public engagement events with researchers and clinicians. The first such event was on 13th November 2018 where they invited people from the database to listen to presentations on diabetes research which they may be interested in. The event was also attended by the charity Diabetes UK and latest information on diabetes and self management was distributed. Members were also allowed to vote on future presentations and make recommendations. The day was highly evaluated and a report was sent to all invited members. The study team also have a twitter account to promote such events. Another large event is being planned in the autumn of 2019. The study team also intend to email and post a twice yearly newsletter giving latest information on diabetes research and other relevant material to members. The first newsletter is anticipated to be in Spring 2019. It is the study team's intention to submit abstracts and presentations for relevant diabetes professional conferences e.g. Diabetes UK and International Diabetes Federation and publish our findings in appropriate journals e.g. Diabetic Medicine. All outputs will be aggregated with small numbers suppressed in line with the HES Analysis Guide.

Processing:

No contact will be made with any individual(s) who could be identified from the information supplied, other than as specified in the protocol and associated documents. Use of the data supplied is for the sole purpose set out above. The Data must not be shared with any other organisation or named individual not explicitly referred to within this agreement. If the information referred to herein is subject to an FOI or other request to share the Data, then agreement from the HSCIC must be sought before undertaking this. The Dataset must not be shared with any third party in the format in which it is provided to you by the HSCIC. Information tools derived from this Dataset will not be provided to any organisations without the specific consent of the HSCIC. Any publications derived from this Data by any party must be subject to ONS confidentiality guidance on the release of birth and death and Health Statistics: The National Statistician sets standards for protecting confidentiality, including a guarantee that no statistics will be produced that are likely to identify an individual unless specifically agreed with them, where the guarantee is judged against the standard that ‘it would take a disproportionate amount of time, effort and expertise for an intruder to identify a statistical unit to others, or to reveal information about that unit which is not already I the public domain.’ Specifically, undertake to ensure that appropriate controls are in place, to ensure compliance with the HSCIC’s, Small Numbers Special Terms and Conditions. Such controls will, as a minimum, meet the requirements of condition 3.3 of the Small Numbers Special Terms and Conditions and more generally satisfy Section 5 of the ONS confidentiality guidance.


Project 5 — DARS-NIC-25945-T8Q0Z

Opt outs honoured: N

Sensitive: Sensitive

When: 2016/09 — 2017/02.

Repeats: Ongoing

Legal basis: Informed Patient consent to permit the receipt, processing and release of data by the HSCIC

Categories: Identifiable

Datasets:

  • MRIS - Flagging Current Status Report
  • Hospital Episode Statistics Admitted Patient Care
  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Outpatients
  • MRIS - Cohort Event Notification Report

Objectives:

The BEST2 study was set up to investigate the safety and performance of the Cytosponge™ test for diagnosing Barrett's Oesophagus (BE) over three years. It is a case-control study, participants (1344 men/women aged 18-60 years) are either patients with known BE (860) or controls individuals' with reflux or indigestion (dyspepsia) symptoms referred for endoscopy (484). All participants swallowed a Cytosponge™ device prior to endoscopy, which is processed for a number of different biomarkers. Results are then compared with endoscopy findings. After the first year of study all cases completed Cytosponge™ and endoscopy procedures. The BEST2 study would like to follow-up participants using HES information in order to collect long-term efficacy and safety data. The main objective is to assess the reproducibility and efficiency of the Cytosponge™ procedure and results to determine the risk of cancer progression, and to compare with endoscopic biopsies, for potential use in the NHS or other health care setting. The Cytosponge™ test showed a specificity of 92% and a sensitivity of around 80% which increases with segment length and is not compromised in the presence of dysplasia. The Cytosponge™-TFF3 test could be used to diagnose patients who would otherwise be referred to endoscopy to rule out BE and therefore be a more systematic screening test for use in primary care. Follow up of all participants is needed to continue monitoring the presence of dysplasia and any risks of cancer progression. Currently guidelines recommend endoscopic screening for BE in individuals with multiple risk factors. However, the cost advantages associated with the Cytosponge™-TFF3 test means screenings could be offered to a larger number, thereby improving the potential to correctly identify a greater amount of newly diagnosed BE cases. The BEST2 study trial site (Cambridge) propose to provide the identifiable data to HSCIC, for all participants that have consented to this, for linkage. HSCIC will return the linked data set with any patient identifiable information (NHS number, names etc) that Cambridge previously sent to the HSCIC along with requested data sets. Identifiable data is requested so that Cambridge can update any new information on the cohort in order to contact participants. Cambridge will not contact the patients directly but will inform the participants GP who in turn will discuss with the participant. Cambridge will then send a patient level pseudonymised version of this data set to QMUL for further processing linked using a unique study ID. Data linkage will improve the quality and integrity of data already collected. All clinical data received, will be stored in a distinctly separate database to the patient identifiable data. The data will not be used for any purpose other than to meet objectives as stated in the trial protocol & will not be shared with any third party organisation. Any information which is used for publication in peer reviewed journals will be anonymised (i.e. aggregated data) & not presented at the individual level.

Expected Benefits:

The BEST2 study recruited 1344 men and women within the UK with either known Barrett's Oesophagus (cases) or individuals with reflux or indigestion symptoms referred for endoscopy (controls). All participants were screened with the Cytosponge™ device to test its potential in determining the risk of cancer progression (in conjunction with biomarkers of risk). The results were then compared to routine endoscopy findings. The study objectives are as follows: Primary objectives QMUL are interested in obtaining the data from these patients to link the biomarker work performed on the samples collected and the risk of progression to cancer. Performance and safety characteristics of the Cytosponge™ test Effectiveness of the Cytosponge™ for diagnosing BE compared with endoscopy, including specificity (from controls) and sensitivity (from cases) For patients with BE, the ability of Cytosponge™ biomarkers to risk stratify patients, according to their future cancer risk in comparison with the dysplasia grade obtained from endoscopic biopsies. Secondary objectives Differential sensitivity of screening BE with dysplasia (low and high grade) compared to non-dysplastic BE. Determine the reproducibility of the Cytosponge™ result by repeated testing in a subset of individuals Logistics of high-throughput sample processing and automated analysis of Cytosponge™ specimens for use in routine NHS or other health care settings. Surplus material will be used for testing emerging biomarkers. Analysis of NHS records will help measure long term effectiveness of the Cytosponge biomarkers to risk stratify patients. In addition to identifying its efficacy in an NHS setting as an alternative to endoscopies. The potential benefits of this research include reducing the use of invasive procedures for patients and by providing evidence for policy decision makers for the use of a safe, minimally invasive, cheaper, and easily administered method to diagnose and screen for this condition A target date has not yet been assessed due to the unknown level of data expected to be received as processing time cannot be estimated yet.

Outputs:

Data outputs produced will not contain any patient identifiable data or be shared with any third party organisation. The only outputs produced will be publications for research purposes using aggregated data with small number suppression in line with the HES Analysis Guide. At this point in time, QMUL cannot comment on the name of journal or conferences as it will depend on the result. Possible journals include PLOS Medicine, Gastroenterology or Gut. The conferences would be Digestive Disease Week and/or United European Gastroenterology Week. It is not yet possible to give a time frame as the length of time to obtain the data has already extended longer than anticipated but once data has been received QMUL expect processing to be completed in around 4-6 months.

Processing:

The data will be processed jointly by Cambridge university and the data controller QMUL. Cambridge university will send the cohort data to the HSCIC for data linkage. Once returned from the HSCIC Cambridge will integrate the data into their database updating any new patient information. Then the identifiable information received from the HSCIC will be removed by Cambridge and this anonymised data set will be sent (via secure encrypted data transfer) to QMUL linked via a unique identifier. QMUL will use this data to link baseline Cytosponge™ findings to disease progression. The following outcomes will be monitored; Barrett’s oseophagus, Barrett’s high grade dysplasia, adenocarcinoma of the oesophagus, squamous cell carcinoma of the oesophagus and oesophageal cancer. All data received at the QMUL Centre for Cancer Prevention, (CCP), are stored electronically on an Oracle database (Oracle 11g). The system is self-contained within the Barts Cancer Research Centre (QMUL) network, within the Queen Mary University of London (QMUL) network, but firewalled off from the rest of the network, in addition to external connections.


Project 6 — DARS-NIC-245768-V0N2T

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

Sensitive: Sensitive

When: 2021/01 — 2021/02.

Repeats: One-Off

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

Categories: Identifiable

Datasets:

  • Cancer Registration Data
  • Civil Registration - Deaths
  • Demographics

Objectives:

Prostate cancer is the most common cancer affecting men in the UK with over 47,000 cases in 2015, representing over a quarter of all new cancer cases in males, and was the second leading cause of cancer death in 2015. Despite its rate of incidence and mortality, men are more likely to live with prostate cancer, as overall more than 85% of men diagnosed with prostate cancer will survive for ten or more years in England and Wales. Localised prostate cancer in particular presents unique challenges in terms of differential diagnosis, and prognosis of aggressive versus indolent cancer and optimum treatment of clinically localised prostate cancer is poorly defined. Current inability to make the distinction poses the greatest problem when deciding appropriate clinical management strategies. This inability to biologically distinguish indolent from aggressive prostate cancer has led to over-diagnosis (identification of false positives) and overtreatment (treating false positives) of prostate cancer in cases where the cancers would not have caused clinical consequences. While only half of patients have potentially high-risk prostate cancer, up to 90% of patients with low-risk prostate cancer still undergo radical prostatectomy. It is known that cancer stage at diagnosis strongly influences the choice of treatment and anxiety also impacts on treatment decisions, with men opting for active treatment rather than active management. Overtreatment of low and intermediate risk prostate cancers is therefore a concern, especially in view of recent research which has shown there is no superiority of active treatment over active management in terms of survival. Traditionally, standard clinical factors such as Prostate-Specific Antigen (PSA), Gleason Grade (GG) and clinical stage have been used to predict clinical tumour behaviour and stratify risk for progression. In particular, PSA testing is widely used as a screening tool for prostate cancer and to determine clinical management, despite its poor specificity and positive predictive value at borderline PSA values. Although PSA can be evaluated in conjunction with GG, tumour stage and other factors to produce the Cancer of the Prostate Risk Assessment (CAPRA), there remains significant heterogeneity in outcomes within these standard groupings and the aforementioned factors are not sufficiently discriminating. Biomarkers have the potential to discriminate between low and high-risk localised prostate cancer. They play a significant role in oncology and are a means to better understand the molecular mechanisms behind tumour growth and progression. As defined by the National Institutes of Health Biomarkers Definitions Working Group, a biomarker “is a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention.” Biomarkers can be detected in blood, urine and tissue samples and can measure macro-molecules, such as DNA and RNA, cells and other biological processes. Biomarkers are critical in the field of prostate cancer as they can determine diagnosis and prognosis, as well as recurrence and treatment response. Novel biological and pathological markers are urgently required in order to develop more accurate prognostic models that can be routinely used in the clinical setting. These models would significantly improve clinical management strategies and patients’ decisions on diagnosis with localised prostate cancer, avoiding interventions with significant morbidity and toxicity. The Trans-Atlantic Prostate Group (TAPG) was established to examine the hypothesis that through a detailed retrospective analysis of outcome in a group of men with clinically localised prostate cancer at diagnosis, variables such as biological, pathological and clinical markers, could be identified that might accurately predict the prognosis of clinically localised prostate cancer. In 1999, the TAPG group initiated the “Prognostic Factors in Prostate Cancer for Patients Treated by Watchful Waiting” study, referred to as the TAPG study. It is a retrospective population-based tissue sample study in men diagnosed with localised prostate cancer 1990-2006, inclusively. The study began as a collaboration between Queen Mary University of London (QMUL) and Sloan Kettering hospital in New York where most biomarkers analyses were performed. Hence the study was named as “Trans-Atlantic” (and the name shortened to “TAPG” for convenience), but the recruitment of prostate cancer patients came from the UK only. Initially the cohort comprised men diagnosed with prostate cancer with transurethral resection of the prostate (TURP) and needle biopsies 1990-1996, but was expanded in 2014 to include men diagnosed with prostate cancer 1990 – 2006. Data was collected from six regional cancer registries and eligibility was confirmed via hospital sites, which sent the relevant tissue samples to the TAPG Central Coordinating Office (CCO). Selection of eligible patients for the study completed in 2010. Since that year the TAPG CCO has been collecting cancer registration and mortality updates on the cohort members from regional cancer registries. The Sloan Kettering Hospital is now no longer involved in this study. It was involved in the early stage of the biomarker analysis phase of the TAPG study, but this work is now being carried in its entirety at QMUL. The CCO at the Centre for Cancer Prevention (CCP) at Queen Mary University London (QMUL) seeks to continue to collect cancer and mortality data provided by NHS Digital. seeks to continue to collect cancer and mortality data via data provided by NHS Digital. The data sets returned by NHS Digital will provide the TAPG CCO with vital information on prostate cancer progression. These data sets will be linked, via a study ID and personal identifiable data, to the participants' pathology data already collected from the hospital sites at which they were diagnosed, as well as clinical data already provided from the now dissolved regional cancer registries. The reports received from NHS Digital are essential in enabling the CCO to determine which pathological, clinical and biological markers can differentiate localised, indolent cancers from more aggressive cancers. The analyses will also enable the research community and patients to make more informed decisions on treatment pathways for prostate cancer, potentially avoiding unnecessary, highly invasive and toxic radical treatment. Although the study will not directly benefit patients diagnosed with prostate cancer after 2006, the assessment of the prognostic model will improve the decision-making process for men diagnosed with prostate cancer in the future. If implemented clinically, men diagnosed with prostate cancer will have a more accurate risk of progression and will be able to make an informed decision on treatment options available to them and whether or not to undergo radical therapy or embark on Active Surveillance. Lack of use of prognostic factors in routine clinical practice is currently a stumbling block to the over-diagnosis and over-treatment of prostate cancer. The prognostic factors under evaluation as part of this protocol could be integrated into the clinical setting in the future, as they are compatible and complementary to other clinical tests. As stated in the 2014 NICE guidelines on ‘Prostate cancer: diagnosis and management’: “Further research is required into the identification of prognostic indicators in order to differentiate effectively between men who may die with prostate cancer and those who might die from prostate cancer. The greatest uncertainties in managing prostate cancer are around the identification of which cancers are of clinical significance and over the choice of radical treatment, and in which settings they are appropriate. With the diagnosis of prostate cancer being made more frequently in asymptomatic men, it is of growing importance to know which of these men are likely to benefit from aggressive treatment.” https://www.nice.org.uk/guidance/cg175/resources/prostate-cancer-diagnosis-and-management-pdf-35109753913285 Therefore there remains an unmet need in terms of identification of prognostic indicators in differentiating aggressive and indolent cancers. The legal bases for processing this data is Article 6(1)(e) ‘performance of a task carried out in the public interest’ pursuant to the GDPR. As the research involves the use of health data, which is included in the definition of special categories of personal data, it requires an additional condition for processing, namely article 9 (2) j which details that processing is necessary for scientific and research purposes, subject to appropriate safeguards. This study has Health Research Authority (HRA) Confidentiality Advisory Group (CAG) ‘Section 251 support’ (CAG reference: ECC 3-06(m)/2009), which permits the processing of confidential patient information without consent and HRA Ethics approval (REC Ref. 99/3/040).

Yielded Benefits:

YIELDED BENEFITS ACHIEVED PRIOR TO RECEIVING ANY NHS DIGITAL DATA The use of the previously collected Registry data has allowed the research team to identify numerous cases of prostate cancer, which are a part of the TAPG cohort. Collection and processing of data from previous TAPG research has contributed directly to the primary endpoint of the trial (death from prostate cancer) and will continue to provide valuable information on understanding the pathological features of prostate cancer. It was also a useful aid in determining if the prediction of disease recurrence could be improved by the addition of other pathological, biological or molecular markers to the standard clinical factors. The data sets previously received have also helped with observing long term outcomes in men that are on Active Surveillance, thus getting a better understanding of the disease. The Transatlantic Prostate Group has demonstrated the prognostic value of pathological, biological and molecular markers, as evidenced by the numerous publications from 2006 to present. 1) Clinical First analysis of the first cohort of patients diagnosed with localised prostate cancer (n=2,333) from 1990 to 1996, after a 12-year median follow-up period, focussed on traditional clinical markers: Gleason Grade, serum PSA, clinical stage, as well as cancer extent in biopsy (Cuzick et al (2006)). The analysis demonstrated that Gleason score was an important determinant of cancer-specific mortality. PSA level, clinical stage and extent of disease also had predictive value. However, the study highlighted that long-term follow-up of men was necessary as it was not clear whether mortality rate trends plateau or continue to increase 15-20 years following diagnosis. Furthermore, the analysis identified that men in the cohort with an intermediate prognosis require better markers of disease progression, in order to assist in formulating appropriate management of individual men with prostate cancer. 2) Histopathological Histo-pathological analyses of clinical markers have also generated findings of relevance to the disease area. In 2007 Berney et al examined the diagnostic accuracy and concordance with which diagnostic pathologists in the UK applied Gleason grading criteria, together with the likely consequences of the grades assigned to patient management. The objective was to test the hypothesis that the study would reveal a previous broad interpretation of Gleason grading criteria which, when applied to a large group of untreated prostate cancers and standardized according to conventional and internationally agreed criteria, would be a powerful predictor of prognosis. The study demonstrated that there was a significant error rate both in cancer diagnosis and in Gleason grading, with a significant upgrading in most biopsy and TURP specimens: 7.5% of the examined cases were reassigned to a non-malignant diagnosis. The effect of grade reassignment on the diagnosis and management of prostate cancer is considerable. Very few studies have examined the rate of misdiagnosis in prostatic cancer retrospectively, and none in such a large series as that now reported. The study highlighted that over-diagnosis of prostatic cancer is likely to have numerous consequences if the findings of this study are reflected internationally. From an individual patient’s point of view, the consequences are those of unnecessary distress, psychological impact and the significant risks of over-treatment, including hormonal therapy, radiotherapy or even radical prostatectomy. 3) Genetic The Trans-Atlantic Prostate Group has produced analyses reporting the clinical significance of various genetic markers. In 2008 Attard et al reported CI5orf21 as novel fusion partner for ETV1, and the heterogeneity of ETV1 gene arrangements in human prostate cancer. In a separate publication in the same year, Attard et al also identified a novel category of prostate cancers characterised by duplication of the fusion of TMPRSS2 to ERG sequences together with interstitial deletion of 2+Edel. The authors concluded that the aforementioned genetic sequences could be part of a new mechanism-based prognostic classification for assessing the potential future aggressiveness of every human prostate cancer at diagnosis. 4) Prognostic models & scores With regards to prognostic models, the Trans-Atlantic Prostate Group has developed various models and scores for the prediction of progression of prostate cancer and disease-specific survival. In 2008, Kattan et al developed and internally validated the first nomogram for prediction of disease-specific survival, since outcome prediction models for watchful waiting are rare relative to those available for aggressive therapies such as surgery or radiation therapy. The nomogram predicted well with a concordance index of 0.73 and had good calibration. This is an accurate tool for predicting the probability that a man with clinically localized prostate cancer will survive his disease for 120 months if the cancer is not treated with curative intent immediately. The tool should be helpful for patient counselling and clinical trial design. [The full account of studies completed and the list of bibliographies can be found in Supporting Document SD13]

Expected Benefits:

The TAPG study is the largest and most comprehensive study of long-term cause-specific mortality in men managed for prostate cancer by Active Surveillance. The TAPG study is focused on understanding prognosis of early prostate cancer managed by Active Surveillance. There have been more than 30 scientific papers published, but the most influential has been the PROLARIS cell cycle progression score ( Cuzick et al 2011; 2012), which can help to determine which patients with early disease can be safely managed by Active Surveillance and which need more aggressive radical treatment. It is currently the most widely used prognostic test in the United States and has a major impact on patient management, therefore this research enables it to be applied in the UK to benefit the UK health and care system. The datasets returned by NHS Digital will provide vital information on the pathological, clinical and biological markers and how they can differentiate localised, indolent cancers from more aggressive cancers. Upon receipt of the mortality data from NHS Digital, the research team will link this information with the biomarker data from the tissue block analysis to assess further the predictive value of biomarkers in relation to disease progression and death from prostate cancer. The analyses will also enable the research community and patients to make more informed decisions on treatment pathways for prostate cancer, potentially avoiding unnecessary, highly invasive and toxic radical treatment. One of the strongest biomarkers found to date was a cell cycle progression panel based on RNA expression of 31 genes (Cuzick et all, 2011). It has been widely used around the world, especially in the United States, and is currently being evaluated for cost effectiveness in the UK by the National Institute for Health and Care Excellence (NICE). Although the current results from the various biomarker and outcome analyses do not benefit the TAPG cohort per se, the discovery of predictive markers for prostate cancer death and its implementation in the UK Health system will benefit many patients with prostate cancer.

Outputs:

Further outputs and analyses of the TAPG study will be published in peer-reviewed journals, such as the Lancet, Journal of Clinical Oncology, or Cancer Research and Treatment, aimed at oncologists/clinicians, researchers and health professionals working within the prostate cancer field. In addition, the research team aims to present updated long-term follow-up data at national and international prostate cancer conferences. The public will also be made aware of findings through appropriate channels, such as a Patient and Public Involvement event; One such event was hosted in January 2020 where participants were invited in to discuss research they have been a part of, to obtain their views and opinions as to how they feel about the research carried out so far, and also find out more about the current and future work carried out at the CCP. Publications with significant findings may be accompanied by press releases and subsequent news stories. Publications, conference attendance and results are published regularly and are available on www.qmul.ac.uk/wolfson. There is also a public Twitter account where updates will be posted on. (https://twitter.com/qmccp).

Processing:

All organisations party to this agreement must comply with the Data Sharing Framework Contract, including requirements on 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). The Barts Cancer Centre (BCC) is part of the Queen Mary University of London’s medical school; Barts and the London School of Medicine and Dentistry. The data received by NHS Digital which will be received by the Centre of Cancer Prevention (CCP) is a research unit which falls under the BCC. The BCC hosts a secure, walled of network within the QMUL environment so the legal entity/data controller here is QMUL but for processing data, this will be done under the BCC ‘safe haven’. Queen Mary University of London is therefore the sole data controller. The Centre for Cancer Prevention (CCP) at QMUL will send a TAPG cohort data, i.e. a list of TAPG study participants, to NHS Digital for linkage. The cohort to be submitted is 3,500 individuals. The following identifiers will be sent to NHS Digital by the CCP team at QMUL: -Full Name, -Date Of Birth, -NHS Number, -Study ID -Hospital ID -Previous Cancer registry ID -Date of death (if required) -Post code. For Data Minimisation purposes, the cohort to be provided is limited to men aged 18-76 years at the time of diagnosis. Inclusion criteria also states patients must have had a baseline serum PSA level measured before starting any prostate cancer specific treatment and within 6 months of diagnosis. Patients must have been diagnosed between 1990 and 2006 with clinically localized prostate cancer (clinical stage T1-T3, N0 or NX, M0 or MX) prostate cancer, in the judgement of treating physician. There is no direct patient involvement as this is a retrospective study carried out in patients registered on UK regional cancer databases, as having prostate cancer diagnosed between 1990 and 2006 inclusively. The data received by the CCP at QMUL will not be used for any purpose other than to meet objectives as stated in the trial protocol and will not be shared with any other third party or organisation. The data sets returned by NHS Digital to the CCP team at QMUL will contain identifiable data. This is so that the applicant can confirm the accuracy and strength of each linkage for TAPG participant in the study cohort. This ensures that all survival and cancer incidence data are correctly attributed to the each TAPG participant, thus avoiding any errors which would invalidate the study. On receipt of the returned data set from NHS Digital, the CCP at QMUL will link this data set to the data set contained in the applicant's TAPG study database via the full name, date of birth, NHS number and study ID. As mentioned above, the data linkage enables the correct identification of the TAPG patients from the data set sent by NHS Digital. Once the patients are identified, researchers at the CCP at QMUL will then update the participant entries in the TAPG study database with date and cause of death, prostate cancer progression, diagnosis and coding. Following the linkage and participant identification, all identifiers except the study ID will be removed from the data set returned by NHS Digital and stored as a reference for analysis until study completion. Processed data sets with all identifiers removed except the study ID will be retained for analysis until study completion, following which it will be archived for 20 years. The original data set returned by NHS Digital containing identifiable data will be stored on a separate server meeting NHS Digital security requirements, as per DARS Guidance Notes on Security for where Mortality data is involved. The applicant would like to emphasise that personal identifiable data are stored on the secure network and never on the local drives of unencrypted QMUL desktop PCs. Bart's Cancer Centre IT Network creates back-ups of the data stored on its network. The encrypted backup tapes are stored offsite with Iron Mountain UK Ltd. The data is not accessible by any Iron Mountain employee. All personal identifiable data received at the Centre for Cancer Prevention are stored electronically on an Oracle database, local to and administered by substantive employees of QMUL only. Personal identifiers of study participants are encrypted in the Oracle database and stored separately to the clinical data, with access strictly restricted to only QMUL substantive members of staff who have accredited researcher status. Only these members of staff will have access to the personal identifiable data. Additionally, access to the personal identifiers on the Oracle database is controlled by separate username and password access and is controlled by the IT Department. Staff that are delegated duties such as data entry/cleaning and manipulation are required to undergo an ONS (Office for National Statistics) accredited researcher training to access and view patient identifiable data sets. This training course covers a variety of topics including data security. Upon completion of training and achieving a pass in the test, an accreditation number is provided and this allows the individual to access such data. Accredited researcher training is mandatory to view such data. Staff are strongly advised to complete an information governance training course and upon completion and passing the test, a certificate is provided. The separate mortality and HES servers are self-contained within the Bart’s Cancer Centre IT network, within the QMUL network. They are firewalled from external connections and the rest of the network. All traffic through the checkpoint firewall is logged. Data will be accessed within the network using workstations that have a currently supported operating system which includes security patches. No remote access (i.e. through a VPN/RDP connection) is permitted. No mobile devices will be used to access any data. The network is externally scanned on a regular basis. The data sets from NHS Digital will be stored until the data destruction date as per this agreement with NHS Digital. The storage architecture is compliant with the NHS Digital Data Sharing Framework Contract and once QMUL is issued with a Data Destruction notice, the data from all storage including backups can be securely and permanently removed within 14 days. Data can be securely wiped to NHS Digital standards (multi-pass pattern wiped to at least HMG S5 Enhanced on site and if end of life, degaussed and physically destroyed).


Project 7 — DARS-NIC-15741-J6Y4L

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

Sensitive: Sensitive

When: 2019/12 — 2019/12.

Repeats: One-Off

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

Categories: Anonymised - ICO code compliant

Datasets:

  • Civil Registration - Deaths

Objectives:

Queen Mary University (QMUL) requires mortality data for use in the study ‘A study to investigate the association between selective uptake of cervical cancer screening and all-cause mortality’. The study was designed and will be conducted by researchers at Queen Mary University of London with the aim of understanding whether women who suffer health problems are less likely to attend cervical cancer screening. The aim of the cervical cancer screening programme is to detect and treat precancerous lesions to avoid the development of cervical cancer. This results in cervical cancer being observed less frequently in women who have attended cervical screening. However it is not clear whether the low risk of being diagnosed with cervical cancer following a normal screening test is due to the screening test alone or whether it also reflects the possibility that those women who attend screening are healthier that those who do not, i.e. they are already at lower risk of cervical cancer than women who do not attend. To explore whether this phenomenon exists, and its magnitude, the researchers have set up a case/cohort record linkage study involving the screening history of two samples of women. Phase 1 is a sample of women who died between 1992 and 2012 (and were aged 20-69 in 1992) who are described as ‘cases’. Phase 2 is a sample of women who have the same month and year of birth as the women identified in the 'cases', which is described as the ‘cohort’; it is possible that some of the individuals selected randomly into the 'cohort' will be deceased, therefore the 'cohort' will have a mixture of patients who are alive and deceased. The study was designed and will be conducted by researchers at Queen Mary University of London with the aim of understanding whether healthy women are more likely to engage with cervical screening and, if they are, to establish the magnitude of this effect. The study requires linkage of Civil Registrations mortality data to the cervical cancer screening database (also known as NHAIS). The maximum estimated sample size of the 'cases' is n=26,850 - it is likely to be less than this because of overlaps in the specific mortality-cause samples and the all-cause mortality sample. There will be an equal number of women identified by NHAIS in the ‘cohort’. Therefore the maximum sample size would be n=53,700 - it is likely to be ca. 50,000, both because of the overlap within the 'cases' as described above, and because it is possible that some of the randomly-selected 'cohort' women will also be in the 'cases' group. NHS Digital will be providing the Civil Registrations mortality data (under this DSA) and NHS Digital will be providing the NHAIS cervical screening data (under a separate DSA with the Data Controllers NHS England and PHE). QMUL will receive only data that is pseudonymised, and where records have been removed for anyone who has registered a national data opt-out. QMUL will link the pseudonymised datasets using a study unique ID. QMUL will analyse and report the data. Queen Mary researchers applied for and secured funding from Cancer Research UK to undertake this work. Cancer Research UK are not involved in processing data or controlling the aims or directions of this study. The aim of the work is to assess whether those that do not engage with screening are more likely to die (from all-cause mortality and from a list of pre-specified cause specific codes). It has always been hypothesised (but never proven) that women in poor health are less likely to attend cervical screening (and hence more likely to die). Queen Mary researchers aim to find out if this is true and to quantify the magnitude of this effect. The ultimate aim is to use the findings to improve the cervical screening programme. This is a one-off new study. It is not part of a bigger study, although one other similar study which used Dutch data exists. The easiest way to find women in poor health is to select women who have died and compare them to women of the same age (at the time of death) who are still alive. Hence, mortality data from England is required, as are cervical screening histories from NHAIS to obtain screening histories for them. By comparing screening histories between both groups of women we can estimate the risk of death (overall and cause specific) among those who do and do not attend cervical screening. Hence, QMUL would like the following information on women identified by Civil Registrations data: - Month and Year of Death - Cause of death (ICD-10)

Expected Benefits:

This research will produce the following benefits to academics studying the effectiveness of screening: inform as to whether current estimates on the effectiveness of cervical cancer screening are overly optimistic and require adjustment, and understanding the role of self-selection in screening; this will enable academics to improve screening programmes. Having a better understanding of health inequalities will enable policy makers to reduce inequalities. The benefits should be observed as soon as the outputs are in the public domain (i.e. once they are published or presented at a conference/meeting) around August 2019, although this will be dependent on the date QMUL receive the data. The level of data contained in the outputs from this study will always be aggregate data with small numbers suppressed, in line with the HES Analysis Guide.. The study will produce a comprehensive report on the analysis and results from the data received from NHS Digital - results for publication in a peer review journal will be produced within 9 months of receiving the data. Additionally, the results will be presented to Public Health England at their yearly Cancer Outcomes conference, and directly to the programme leads through the management group for the Audit of invasive cervical cancer (see below). The Cancer Outcomes conference is held every year in or around June (http://www.datascience.manchester.ac.uk/events-1/events/phe-cancer-services-data-and-outcomes-conference-2018/). The research team at QMUL has a yearly meeting with the programme leads at PHE through the management group for the case-control study. A piece on the results from this study will be written for the study cancer prevention blog https://blogs.kcl.ac.uk/cancerprevention/ and results will be promoted through the project's twitter feed. The research will also be submited to at least one international conference (either Eurogin or International Papilloma Virus conference (IPV)). The results will be used to adjust statistical outputs from other studies of cervical screening being carried out by the team (and once results are published by other teams working in cervical screening). However, the data is not intended to be used in the development of algorithms, the testing and development of tools or in new technologies. For over a decade the research group at Queen Mary has been evaluating the cervical screening programme on behalf of Public Health England. The research group runs a case-control study which includes 90% of women diagnosed with cervical cancer and a number of women invited for screening but who do not have cancer. Feedback to the service provider and all of the staff who contribute to the success of the programme is done through an annual report. https://www.qmul.ac.uk/wolfson/research-reports/centre-for-cancer-prevention/nhscsp-audit-of-invasive-cervical-cancer/. Hence, the researchers at QMUL have established a very strong collaboration with the cervical screening programme.

Outputs:

An academic paper will be published in a cancer specialist open-access, peer-reviewed journals on risk of death (all cause and cause specific mortality) by attendance to screening by age and screening period. A simplified version of the findings will be disseminated to charities (Cancer Research UK and Jo's Cervical Cancer Trust [www.jostrust.org.uk]) and published on the Wolfson Institute website. Based on the publication a short presentation will be developed to summarise the findings for a range of stakeholders, including healthcare professionals and cervical screening policy makers. Findings will be presented at the following events: Public Health England patient outcomes conference, Human Papilloma Virus conference and Eurogin conference. A close working relationship between researchers at QMUL and cervical screening policy makers already exists. The aim is to complete the study 6 month after the data has been released to QMUL and publish outputs 3 months later. The aim is to deliver the peer review manuscript by August 2019, but achieving this will depend on when the data is released. Output will be published as aggregated level data with small number suppressed, in line with the HES Analysis Guide.

Processing:

The study design is a case/cohort. The cases will be identified in phase 1 and the cohort in phase 2. *****Data flows in Phase 1 – ‘Cases’ **** Step 1. NHS Digital will identify a sample of ‘cases’ (i.e. a sample of deaths between 1992 and 2012 in England (from Civil Registrations (Deaths) dataset) among females born between 1923 and 1972 (and therefore aged 20-69 in 1992) using specific ICD10 causes of death as specified in the Additional Information section, and apply opt-outs. This will be a maximum of 26,850. a) NHS Digital securely transfers a file of identifiers (NHS number + DOB + Case PseudoID) to NHAIS. b) NHS Digital securely transfers a de-identified file (Case PseudoID + ICD-10 cause of death + month & year of death) to researchers at QMUL. Step 2. NHAIS identifies screening histories of individuals within the file from NHS Digital. c) NHAIS securely transfers a de-identified file (Case PseudoID + month/year of screening test) to QMUL researchers *****Data flows in Phase 2 - ‘Cohort’***** NHAIS will identify one woman at random for each case from the cervical screening (NHAIS) database to have the same month and year of birth as each ‘case’ woman - this will constitute the cohort. Note that a 'cohort' member can also be in the ‘case’ sample. Step 3. Once the ‘cohort’ is identified NHAIS will: d) securely transfer a file of identifiers (NHS number + DOB + Control PseudoID) to NHS Digital. e) securely transfer a de-identified file (Control PseudoID + month & year of screening test) to QMUL Researchers Step 4. NHS Digital identifies any deaths among the ‘cohort’ file provided by NHAIS and applies opt-outs. f) NHS Digital securely transfers a de-identified file (Control PseudoID + cause of death + month/year of death+ strength of the match (pass 1 to 8)) to QMUL researchers Data flows are designed to minimise the data shared between NHS Digital and NHAIS (as well as that shared with QMUL). The direct flow of screening history data from NHAIS to QMUL is covered by a separate Data Sharing Agreement (DSA) (see SD6). The direct flow of mortality data from NHS Digital to QMUL is covered by this DSA. As the two datasets and relevant flows are covered by separate dissemination arrangements the NHS Digital and NHAIS data production teams are content with the data dissemination model detailed below. Hence the end user files consist of four separate flows made up of: In Phase 1 ‘Cases’ PseudoFile 1 – contains only screening histories with a unique case ID (sent by NHAIS) PseudoFile2 - contains only mortality data with a unique case ID (sent by NHS Digital) In Phase 2 ‘Cohort’ PesudoFile 3 – contains mortality outcomes for ‘cohort’ with a unique cohort ID (sent by NHS Digital) PesudoFile 4 – contains screening histories for ‘cohort’ with a unique cohort ID (sent by NHAIS) Researchers at Queen Mary will receive a total of 4 pseudonymised files identified which can be linked by unique case/cohort ID. These separate flows will allow the researchers to identify the ‘cases’ and the ‘cohort’ samples. *********************************** Data will be stored securely at the Wolfson Institute of Preventive Medicine, QMUL. Charterhouse Square, London EC1M 6BQ which can be remotely accessed via Citrix receiver or physically at Charterhouse Square. Data accessed remotely cannot be copied or emailed. Data will only be accessed by individuals within the Wolfson Institute of Preventive Medicine who have authorisation by the Institute to access the data for the purpose described, all of whom are substantive employees of Queen Mary University of London. The 4 data files received will be linked and analysed by researchers from Queen Mary University for the purpose specified here. Data will not be linked with any other data. There will be no requirement nor attempt to re-identify individuals from the data. The data will not be made available to any third parties. The study design will be a case-cohort study. The case-cohort design is most useful in analysing time to failure (here mortality) in a large cohort in which failure is rare (mortality from cervical cancer and other HPV related cancers is rare). This is why QMUL need national data rather than regional data. Screening history information will be collected from all deaths (cases) as well as from a sample (cohort) of women registered in NHAIS with the same month and year of birth as cases. Sampling is done without respect to time or disease status, and, therefore, the design is more flexible than a nested case-control design. Simplistically, one can use the cohort to study all-cause mortality and the case-cohort to study cause-specific mortality. QMUL are aware that requesting large amounts of detailed data for women can sometimes be challenging as there is great concern for personal data safety. QMUL have selected this study design in an attempt to reduce the number of records necessary while still ensuring sufficient power to observe a difference in mortality between groups. QMUL have additionally applied several filters to ensure they obtain women eligible for screening before death. QMUL have taken great care to only request the data items that are essential for the study and to minimise the identifiable data requested. 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).


Project 8 — DARS-NIC-147843-8NKTW

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

Sensitive: Sensitive

When: 2016/04 (or before) — 2021/03.

Repeats: Ongoing

Legal basis: Section 251 approval is in place for the flow of identifiable data, Health and Social Care Act 2012 – s261(7)

Categories: Identifiable

Datasets:

  • MRIS - Cause of Death Report
  • MRIS - Cohort Event Notification Report
  • MRIS - Scottish NHS / Registration
  • Civil Registration - Deaths
  • Demographics
  • Cancer Registration Data

Objectives:

The aim of the trial is to determine whether healthy people should be screened and treated for H Pylori infection.

Yielded Benefits:

No outputs have been produced as the data analysis plan was designed such that no analysis would be undertaken until sufficient numbers of stomach cancers had occurred. The numbers reported are not sufficient yet.

Expected Benefits:

If the trial determines that screening and subsequent eradication of H Pylori reduces the incidence of stomach cancer this will have an enormous benefit to the whole population in the UK and worldwide. Both screening and eradication (a 1 week course of antibiotics) are simple and cheap. Around 40% of people have H Pylori infection. In 2015 6740 people developed stomach cancer. If screening does work and prevents 20% of stomach cancers then this will mean that over 1,300 stomach cancers will be prevented. With 4 out of 10 people in the UK thought to have H Pylori infection potentially 40% of the population could have their risk of stomach cancer reduced. The extension of our data sharing agreement will allow us to obtain enough cases of stomach cancer to provide the information needed on the value of screening. This trial will provide evidence concerning whether screening for HPylori infection is worthwhile. If the results from the trial are positive then we would be in a position to contact PHE to discuss implementing a population screening programme.

Outputs:

The following outputs will be produced : A peer review paper analysing the results from the trial will be submitted to an open-access peer reviewed journal within one year after sufficient stomach cancers have occurred.This paper should be influential in deciding whether to screen for H Pylori infection . The final report of results will be submitted to CRUK. This will cover all findings of the study. For each paper published, a short presentation may be developed to summarise the findings for a range of stakeholders, including healthcare professionals and patient groups. The study website will provide links to the open access papers and will offer free downloads of accessible summaries of findings. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide/compliant with the MHSDS disclosure control rules including suppression and rounding.

Processing:

All 62,454 participants in the HPSS have been flagged at NHS Digital. Information on deaths and cancer registrations is requested to be received every 6 months electronically. The electronic information will be downloaded onto a secure server, protected by a fire-wall, based in the Wolfson Institute of Preventive Medicine. The data are then merged with the HPSS study database by the database manager in the Wolfson Institute. The data are stored on the server, with any identifiers stored separately from the clinical information. The database manager provides the study statistician with pseudonymised information on the numbers of deaths and cancer registrations that have occurred since the start of the trial. Data will only be accessed by individuals within the Centre for Environmental and Preventive Medicine who have authorisation from the PI (Sir Nicholas Wald) to access the data for the purpose described, all of whom are substantive employees of QMUL. The core dataset will only be accessed by the data manager within the Wolfson Institute. They will produce subsets of the data that will be accessed by the study statistician. Any other person seeking access toi a subset of the data will have to submit a formal reuest to the PI (Sir Nicholas Wald) and justify from a scientific basis all requested information. 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 of QMUL situated in the Centre for Environmental and Preventive medicine who may have access to that data). No further linkage will be performed. The 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. Data is only requested for those participants in the HPSS study. Some identifiers are necessary to ensure that the correct match with the study participant is made. The BUPA identifiers which were used in this study were not totally unique, causing manual checks to be made when any linkage is performed. No outputs have been produced as the data analysis plan was designed such that no analysis would be undertaken until sufficient numbers of stomach cancers had occurred. The numbers reported are not sufficient yet.


Project 9 — DARS-NIC-147747-KRTQ8

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

Sensitive: Sensitive

When: 2020/10 — 2021/04.

Repeats: Ongoing

Legal basis: National Health Service Act 2006 - s251 - 'Control of patient information'.

Categories: Identifiable

Datasets:

  • Demographics
  • Civil Registration - Deaths
  • Cancer Registration Data

Objectives:

A subgroup of the Policy Research Unit team at Queen Mary University of London (QMUL) require the baseline and screening exposure data of the breast screening cohort study (currently already held at QMUL) with follow-up data for cancer incidence and death (currently held to 2005, requesting to 2016). This is for the purpose of evaluating the breast cancer screening programme in terms of its effect on breast cancer mortality and in terms of harms of screening, notably overdiagnosis (identification of false positives). This work was instigated by the Institute for Cancer Research (ICR), Sutton, Surrey, to establish the effect of the NHS Breast Screening Programme on incidence of and mortality from breast cancer. This work is research in the public interest because its purpose is to evaluate the breast cancer screening programme in terms of its effect on mortality and the potential harms due to identification of false positives. Notably, the number of people who would benefit from this research is exceptionally high: - Two million women undergo breast cancer screening every year - One person is diagnosed with breast cancer in the UK every 10 minutes - One in eight women will develop breast cancer in their lifetime - Approximately 11,500 people – overwhelmingly women – die of breast cancer every year Once received from NHS Digital, the data will only be used in identifiable form for the specific purpose of data linkage and will then be pseudonymised. The cohort comprises approximately 2.7 million women born between 1923 and 1945 resident in the study area who were eligible for at least one invitation by the NHSBSP between January 1988 and December 1994. The study area covered 22 Local Health Boards (LHBs) in Wales and 137 PCTs in England, and the study therefore included approximately 38% of the population of England and Wales who were eligible for invitation to NHS breast screening between these dates. Following the closure of the Cancer Screening Evaluation Unit at ICR and the establishment of the Policy Research Unit in Cancer Awareness, Screening and Early Diagnosis (PRU) by the Department of Health (now the Department of Health and Social Care), this work was moved to QMUL, the home institute of the PRU director. The purpose of the work remains the same, to evaluate the major benefits and harms of the NHS Breast Screening Programme. QMUL is the only organisation which will have access to record level data. The ICR is no longer involved in this project. Thus QMUL will be responsible for both data control and data processing. The Department of Health and Social Care established the PRU, led by the applying organisation, QMUL, with a remit among other tasks to evaluate the NHS Cancer Screening Programmes, in terms of both positive and negative effects of the programmes. The primary aims of this programme are to quantify in both relative and absolute terms: (i) The reduction in breast cancer mortality associated with the policy of offering mammographic screening from breast cancer. (ii) The corresponding reduction in breast cancer mortality associated with individual participation in the NHS Breast Screening Programme. (iii) The risk of overdiagnosis (identification of false positives) associated with participation in the programme. Secondary aims include the estimation of self-selection biases, including development of methods of estimation of such, to improve the accuracy of aims (i-iii) above. It is also planned to estimate benefits and harms within narrow age subgroups, and estimate effects of time since screen, to inform policy on the target population and on the frequency of screening. The PRU shall also examine effects by area, to flag up areas where the screening has not worked as well, which will trigger investigation of the cause, and subsequent rectification, if necessary. This is a unique opportunity. Area-based process measures (screen detection rates, false positive rates etc) have long been available but there has been no data on the effect on clinical outcomes by area. This study has been running for more than three decades, it is a stand-alone cohort study, the data has been supplied before, and the objectives remain the same (as noted above). The data required for the aims listed above are (1) Retention of the baseline demographic and screening data which are already held at QMUL; (2) The death and cancer notification data up to 2005, which are also already held at QMUL; and (3) further death and cancer notification data to 2016. These data are requested as they will give the opportunity to estimate the effects of the NHS Breast Screening Programme on mortality from and incidence of breast cancer, with rigorous adjustment for potential biases. In particular, the further notification data to 2016 will give valuable information on the long-term effects. Further background to and details of objectives The NHS Breast Screening Programme (NHSBSP) was introduced in England and Wales in 1988, following the publication of randomised controlled trials (RCTs) and population demonstration projects which reported that mammographic screening could reduce breast cancer mortality by 20-30%. Mechanisms to monitor the NHSBSP in terms of quality assurance were built into the programme and using performance indicators the programme currently appears to be performing well compared with RCTs. However, monitoring the performance of the service does not directly address questions about its effect on breast cancer mortality and the question remains whether the reduced breast cancer mortality observed in RCTs can be achieved in a population programme, such as the NHSBSP. In addition, mortality from breast cancer is falling in the United Kingdom and there is debate as to how much of the reduction is due to screening or improvements in treatment (or some other factor). The difficulty of producing quantitative estimates of the effect of a national screening programme on mortality is well recognised. Population level data have been used to estimate the effect of the NHSBSP on breast cancer mortality in England and Wales, but ecological studies suffer the disadvantages of using aggregated data and are complex. The impact of the NHSBSP on breast cancer mortality in the East Anglian region has been estimated using prognostic characteristics of breast cancers to predict breast cancer mortality. However, to assess the programme rigorously the PRU needs to use individual-level data and mortality from breast cancer as the outcome. The NHSBSP now screens more than 1.3 million women per year and has an annual budget of approximately £52 million. The programme has a high public profile, and debate following recent scientific publications has again raised the issue of whether the Breast Screening Programme is an effective use of NHS resources, and underlined the need for proper evaluation of the programme in terms of its effect on breast cancer mortality. The primary aim of this study is to evaluate the impact of the NHS Breast Screening Programme on mortality from breast cancer. The objective is to use individual-level data on a cohort of women to assess breast cancer mortality in relation to screening history; specifically to compare risks in those invited with a contemporaneous group of the same age who were not invited until a later date. Also to assess: a) mortality from breast cancer in relation to attendance for screening. Those who attend for screening do not represent the entire target population of the programme and bias is introduced by factors influencing the uptake of an invitation. However, mortality among women who attend for breast cancer screening is of considerable interest. b) mortality from all causes, all cancers and vascular disease in relation to invitation to and attendance at screening. Previous literature has reported a relationship between breast cancer screening and non-breast cancer deaths, specifically from other neoplasms and from vascular disease. The applicants are now seeking support to flag the remainder of the cohort who have not been flagged previously in order to obtain up-to-date cancer incidence and mortality follow-up to conduct analysis on a longer-term basis. The objectives of this follow-up phase are to further assess the effectiveness of the screening programme by comparing risk of death from breast cancer in women who were invited for screening with those who were not invited in the same period for follow-up over a 25-year period, and wider objectives related to the impact of the screening programme on diagnosis of breast cancer.

Yielded Benefits:

The study has already borne fruit in terms of publication of medium-term estimates of the reduction in mortality and the overdiagnosis - identification of false positives - associated with invitation to screening: Johns LE, Coleman DA, Swerdlow AJ, Moss SM. Effect of population breast screening on breast cancer mortality up to 2005 in England and Wales: an individual-level cohort study. Br J Cancer 2017; 116: 246-52 Johns LE, Swerdlow AJ, Moss SM. Effect of population breast screening on breast cancer mortality to 2005 in England and Wales: A nested case-control study within a cohort of one million women. J Med Screen 2018; 25: 76-81 Results include a substantial reduction in medium term mortality from breast cancer, with little or no overdiagnosis (Johns et al, Br J Cancer 2017; 116: 246-52). However, it would add considerable value to update these results to long-term effects and to extend the population flagged to the entire cohort.

Expected Benefits:

Measurable benefits are precise and accurate estimates of mortality reduction and identification of false positives as mentioned in Outputs. Because of the large size of the study, estimates can be obtained for specific age subgroups and observed intensities of screening. These will enable decisions to be made regarding changes to the programme, for improved efficiency. There will be additional benefits in terms of estimation of other aspects of screening (age at first and last screen, etc). Two perennial problems with screening evaluation are self-selection bias for screening and the fact that exposure and risk status pertains to date of diagnosis, whereas the clinical endpoint occurs at date of death. This cohort design affords opportunities to surmount both of these problems. As a result, decision makers will have quantitative data on long-term benefits and harms of mammography screening, to inform changes (or not) to screening policy and practice. Such changes might include variation in age groups or interscreening intervals. The results will be made available to the National Screening Committee and to the Breast Screening Advisory Group prior to publication, so that decisions can be made in a timely fashion. As noted above the PRU is represented on the National Screening Committee and on the Breast Screening Advisory Group, so there is no chance that the results will not be considered by these bodies. Potential benefits to patients and public included changes to the frequency of screening, with potentially variable frequencies for different age groups, as occurred following the case-control evaluation of the cervical screening programme. The NHS Breast Screening Programme screens more than two million women per year. Thus there is a large number of potential beneficiaries. Benefits for those charged with evaluating screening will include the methodological advances in terms of adjusting for self-selection bias.

Outputs:

This unique dataset will provide estimates of the major benefit (reduction in breast cancer mortality) and harm (overdiagnosis: identification of false positives) conferred by the NHS Breast Screening Programme. It includes subjects recruited at the initiation of the programme in the late 1980s. Despite substantial experimental and observational evidence there remains uncertainty over the magnitude of the benefit of the NHS breast screening programme in terms of breast cancer mortality and the magnitude of the harm in terms of overdiagnosis (identification of false positives). This extremely valuable data resource has the potential to provide answers to these questions with a precision unrivalled by any other study worldwide. The post-2005 data is needed to answer two specific lines of research: - First, there is a need to estimate the long-term benefit of early detection, in terms of breast cancer mortality. For many cancers, if recurrence does not occur within, say, five years, it will not occur at all. This is not the case for breast cancer, which can recur decades after the original diagnosis, and recurrence can prove fatal. It is therefore important to ascertain whether the benefit of early detection holds for later events as well as earlier. - Second, there are conflicting results in terms of early detection, where some trials show the benefit being maintained up to thirty years and others (particularly in younger age groups) show it declining after ten years. Resolving this issue may have implications for other aspects of management than early detection- it may have implications for treatment of breast cancer occurring at earlier ages, for example. The data controller requires baseline data as already present in the dataset held at Queen Mary University of London, plus notifications of breast cancers and all deaths within the cohort, including cause of death. The notifications of deaths and cancers at the moment only extend to 2005. Queen Mary University of London seek approval for reopening this and giving notifications to 2016 - or 2017 if available -, and request that this be part of the new data sharing agreement. In addition, Queen Mary University of London request that flagging be extended to the entire cohort, rather than only those recruited after 1991. This would give a longer-term and larger population incidence-based mortality effect for the screening programme than is available from any other dataset in the world. In addition, it will provide long-term data on incidence (indeed lifetime incidence in many cases), essential for estimation of the identification of false positives unbiased by lead time. Note that substantial additional funding is available for the further flagging and notifications. This work was instigated more than 20 years ago by the Institute of Cancer Research and was transferred to Queen Mary University of London (QMUL) in 2015, following the award of the Department of Health Policy Research Unit in Cancer Awareness, Screening and Early Diagnosis to QMUL, and the closure of the Cancer Screening Evaluation Unit at ICR. The aims of the work remain the same, to quantify the benefits and harms of the NHS Breast Screening Programme. The expected outputs pertain to methodology, knowledge creation in relation to the effect of breast screening, and potential healthcare policy changes. For this first, this dataset affords an opportunity to understand better and in a UK context the phenomenon of self-selection for screening, as it will contain women who have, contemporaneously, not yet been invited (eg, someone recruited to the cohort in 1989 and first invited in 1994 will have spent 5 years not yet invited to screening) and women who have refused (ie, turned down their invitation) screening, that is it can estimate self-selection unconfounded by time, and the potential changes in population behaviour and attitudes over time. In terms of knowledge, it will give estimates of the benefit of screening in terms of breast cancer deaths prevented and harms in terms of identification of false positives to an unprecedented level of precision. This will result in high-impact publications. Further, the size of the data resource will permit detailed analyses of subgroups and tactical issues, such as time between screens and age at screening with the potential to inform policy as to target population, interval between screens, etc. In terms of the findings reaching decision/policy makers, researchers, clinicians, patients and the public, as with all PRU's research, the results are published in peer-reviewed medical journals. The study has already borne fruit in terms of publication of medium-term estimates of the reduction in mortality and the identification of false positives associated with invitation to screening: Johns LE, Coleman DA, Swerdlow AJ, Moss SM. Effect of population breast screening on breast cancer mortality up to 2005 in England and Wales: an individual-level cohort study. Br J Cancer 2017; 116: 246-52 Johns LE, Swerdlow AJ, Moss SM. Effect of population breast screening on breast cancer mortality to 2005 in England and Wales: A nested case-control study within a cohort of one million women. J Med Screen 2018; 25: 76-81 The further follow-up and associated methodological development will enable estimation of the mortality reduction and the overdiagnosis (false positives) associated with actually being screened rather than only the effect of invitation. It will also allow estimation of the long-term effects. Again, results will be published in peer-reviewed medical journals. However, prior to publication, these will be shared with the relevant policy community. The PRU is represented on all the cancer screening advisory groups and on the National Screening Committee. These bodies will be apprised of the results in a timely manner, as will Cancer Research UK, Macmillan Cancer Care and Breast Cancer Now. In addition, the funding body (the Department of Health and Social Care, via NIHR) will be informed of the results prior to publication. The results shall also be posted on the PRU website. PRU will take advice from the PRU patient and public advisory committee as to the most appropriate form of general public disseminations. In addition, the PRU’s scientific advisory committee will be consulted on options for further dissemination. It is estimated that results will be available in 2020. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide. Note that a new Patient and Public Involvement Board has been set up to advise on: - setting the research direction - obtaining wider involvement from all the population this programme serves (ie, not just patients with lived experience of cancer) - increasing public engagement with the appointment of an academic PPI lead and a PPI co-investigator (Bishop Douglas Lewins). The PAB's first meeting was on 14th March 2019.

Processing:

The data requested will flow from NHS Digital to Queen Mary University of London (QMUL). The data are and will continue to be held securely at QMUL and will not be shared with any other person or organisation. The data specifications are long-established and unchanged. The data will not be linked with other datasets. Data will be analysed mainly by log-linear and logistic-linear regression models to ascertain effects of screening on incidence of breast cancer and on mortality from breast cancer and from other causes. The data sharing agreement is requested for three years, as there will inevitably be delays in reopening and updating notifications of cancers and deaths, and because the analysis of a large and complex dataset will take considerable time. The following fields will be supplied by study investigators for matching. Study id NHS Number Date of birth Postcode Please note: Participants forename and last name will not be used for matching. The following fields will be supplied by NHS Digital to QMUL: NHS Number (linkage field) Details of de-reg and re-reg within the NHS (source PDS) Please note: DREG and DRR are both date fields from the PDS database. They contain the Date of first registration onto the Exeter system and Date of re-registration from the Exeter system respectively. Fact of death (source ONS) Date of death (source ONS) ICD10 cause of death (source ONS) Diagnosis of cancer (source NCRAS) Date of diagnosis (source NCRAS) Site of cancer (source NCRAS) Data will only be accessed by individuals within QMUL who have authorisation from the Principal Investigator to access the data for the purpose(s) described, all of whom are substantive employees of QMUL. The data will not be linked with any other data, or any record level data. There will be no requirement nor attempt to re-identify individuals from the data. The 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 request is for the receipt of cancer registration and death notifications from January 2006 to December 2016 for the cohort of women. The data required is national. The aims require that all deaths within the cohort with cause be supplied, and all cancers, invasive and in situ. 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).


Project 10 — DARS-NIC-12629-B4N5K

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

Sensitive: Sensitive, and Non Sensitive

When: 2019/01 — 2020/03.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(7), National Health Service Act 2006 - s251 - 'Control of patient information'.

Categories: Identifiable

Datasets:

  • MRIS - Members and Postings Report
  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Outpatients
  • Hospital Episode Statistics Admitted Patient Care
  • MRIS - Cause of Death Report
  • MRIS - Cohort Event Notification Report

Objectives:

Established in 1992, the IBIS-I Study investigated the efficacy of tamoxifen (a hormonal drug used to prevent breast cancer) versus a placebo drug (taken daily for five years) in terms of reduction of breast cancer incidence in pre and post-menopausal women at high risk of developing breast cancer. It was a double-blind, randomised placebo-controlled trial that recruited 7,154 women internationally (of which 4,277 were UK participants), aged 35-70 years. The primary outcome measure was the incidence of breast cancer, including ductal carcinoma in situ (cancer cells in the lining of the breast milk duct) and side effects present in the patients were also investigated. Recruitment to the study completed in 2001 and the intervention (placebo/tamoxifen) ended in 2007. In early 2008 the Research Ethics Committee (REC) approved the conversion of IBIS-I to an epidemiological cohort study. During 2007–2016 participants were followed-up via an annual postal questionnaire. In 2002, initial results found that tamoxifen reduced the risk of invasive breast cancer by 31%. Mortality from non-breast-cancer causes was not increased by tamoxifen. However, the analysis concluded that the overall risk/benefit ratio for the use of tamoxifen in prevention remained unclear and that continued follow-up of trial participants was essential. A 2007 analysis on long-term tamoxifen prophylaxis for breast cancer confirmed the preventive effect of tamoxifen in terms of breast cancer incidence and that this was constant for the entire follow-up period. No reduction in size of benefit was observed for up to ten years following participant randomisation. Additionally, tamoxifen-related side effects such as thrombo-embolism were not increased anymore after the 5-year treatment period. These results therefore demonstrate that the benefit-to-risk ratio of tamoxifen improves with increasing duration of follow-up. Thus, how much additional benefit will be seen long-term remains an important question. Based on these above-mentioned results, the Central Coordinating Office (CCO) at the Centre for Cancer Prevention (CCP) at Queen Mary University London (QMUL) seek to continue to passively collect cancer registration and mortality data via MRIS products provided by NHS Digital, and to also start collecting Hospital Episode Statistics (HES).The data sets returned by NHS Digital will provide vital information on side effects, survival and breast cancer incidence. These data are essential in enabling the CCO to determine whether tamoxifen continues to have a long-term beneficial impact in terms of breast cancer incidence, survival and side effects after the initial 5-year treatment period. The CCO will be able to perform a thorough risk-benefit assessment of tamoxifen and a 30-year median follow-up analysis. These analyses will enable the research community to make more informed decisions on tamoxifen vis-a-vis patient safety and both short- and long-term effectiveness. The results so far have been very important for breast cancer in the preventive setting.

Yielded Benefits:

Registry data provided to date has identified numerous cases of breast cancers, other cancers and side-effects in participants with whom the Centre for Cancer Prevention at Queen Mary University London have no direct contact and therefore not already known by the study team. Collection of this data has contributed directly to the primary endpoints of the trial and will continue to provide valuable information on the long-term risk/benefit status of the trial intervention. This long term data is essential for informing bodies such as NICE to inform the prescribing utility of these drugs as chemopreventive agents. The CCP use registry reported events as a prompt for further investigation by asking participants' current GP practice to confirm the event and also provide any further information e.g. cancer grade, size, receptor status that is required to meet secondary and exploratory objectives. So far the long-term follow-up has shown that tamoxifen significantly reduces breast cancer incidence in women at risk of developing the disease. The thromboembolic side effects and increase in gynaecological problems (including endometrial cancer) disappear after the active treatment phase has ended. The trial has also provided further information on the impact of tamoxifen on breast density: mammographic density appears to be a prognostic marker for improved long-term survival in participants receiving adjuvant tamoxifen. Notably, the IBIS-I study won the Cancer Research UK Prize for Translational Cancer Research at the 2014 NCRI Cancer Conference for its improvements in cancer research (http://www.ibis-trials.org/thetrials/ibistrials/ibis-i-award). Moreover, NICE updated its guidelines on prescription of tamoxifen to women at high risk of developing breast cancer, a decision partly based on results of the IBIS-I Study (first published June 2013, update August 2015; https://www.nice.org.uk/guidance/CG164 and https://www.nice.org.uk/guidance/cg164/evidence/surveillance-review-decision-november-2015-pdf-2178797581).

Expected Benefits:

The data sets returned by NHS Digital will provide vital information on side effects, survival and breast cancer incidence. IBIS-I is currently the only cancer prevention study to have achieved a 20 year follow-up so far. Using these data sets, the IBIS-I Central Coordinating Office (CCO) will conduct research and analyses that will help provide further information on: • the efficacy of tamoxifen prophylaxis in reduction of the risk of breast cancer in high-risk women (pre- and post-menopausal); • the safety of tamoxifen prophylaxis in reduction of the risk of breast cancer in high-risk women (pre- and post-menopausal); • the long-term all-cause mortality rate related to tamoxifen; • a thorough risk-benefit assessment of tamoxifen. The research will help further inform public health bodies such as NICE on the safety and efficacy of tamoxifen as a chemopreventive agent. These analyses will enable patients and women at risk of developing breast cancer to make a more informed decision on the preventive treatment options available to them. An estimated 15% of women in the UK could benefit from preventive tamoxifen therapy and therefore results from the IBIS-I study are of great relevance to these women. The target date for these benefits is anticipated in 2019 and 2025, in line with the anticipated analyses of the 20 and 25 year follow-ups respectively.

Outputs:

The first results of the IBIS-I have already been published (Cuzick J, et al. The Lancet. 2002;360: 817-24; Cuzick J, et al. J Natl Cancer Inst. 2004;96:621-628; Cuzick J, et al. J Natl Cancer Ins. 2007;99:272-82; Cuzick J, et al. San Antonio Breast Cancer Symposium. 2014, Dec 9-13; Cuzick J, et al. Lancet Oncol. 2015;16(1):67-75). Further outputs and analyses of the IBIS-I trials will be published in peer-reviewed medical journals such as The Lancet, The Lancet Oncology, Breast Cancer Research and Treatment and the British Journal of Cancer with at least two anticipated publication dates in 2020 and 2025 Outputs are anticipated in 2019 when there is 20 years of median follow-up data available. It is further anticipated that Queen Mary University of London will run an updated analysis in 2024 (25 years median follow-up). The public will be made aware of the research progress through two web pages (http://www.ibis-trials.org/thetrials/yourstories/press-articles and http://www.wolfson.qmul.ac.uk/current-projects/ibis-1#publications), and also via updates on Twitter (https://twitter.com/qmccp). The anticipated audiences of the above outputs are researchers, scientists, stakeholders for the wider project and research participants. Policy makers - such as NICE will also be targeted with the outputs of the study. NICE will be informed on the safety and efficacy of tamoxifen as a chemopreventive agent, allowing more information available to women at risk of developing breast cancer on the options available to them. An estimated 15% of women in the UK could benefit from preventive tamoxifen therapy and therefore results from the IBIS-I study are of great relevance to these women. Data contained in any outputs will be aggregated with small numbers suppressed in line with the HES analysis guide.

Processing:

All organisations party to this agreement must comply with the Data Sharing Framework Contract, including requirements on 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). The Centre for Cancer Prevention at QMUL will send IBIS-I cohort data, i.e. a list of IBIS-I study participants, to NHS Digital for linkage. The following identifiers will be sent to NHS Digital by the CCP team at QMUL: -Full Name, -Date Of Birth, -NHS Number, -Postcode, -Study ID. The data received by the Centre for Cancer Prevention at QMUL will not be used for any purpose other than to meet objectives as stated in the trial protocol and will not be shared with any other third party or organisation. The data sets returned by NHS Digital to QMUL will contain identifiable data. This is so that the applicant can confirm the accuracy and strength of each linkage for IBIS-I participant in the study cohort. This ensures that all side effects, survival and cancer incidence data are correctly attributed to the each IBIS-I participant, thus avoiding any errors which would invalidate the study. On receipt of the returned data set from NHS Digital, the Centre for Cancer Prevention at QMUL will link this data set to the data set contained in the applicant's IBIS-I study database via the full name, date of birth, NHS number and study ID. As mentioned, the data linkage enables the correct identification of the IBIS-I patients from the data set sent by NHS Digital. Once the patients are identified, researchers at QMUL will then update the participant entries in the IBIS-I study database with date and cause of death, cancer recurrence, diagnosis and coding, and any HES data relevant to the study (fractures, cardiovascular and thrombo-embolic events). Following the linkage and participant identification, all identifiers except the study ID will be removed from the data set returned by NHS Digital and stored as a reference for analysis until study completion. Following receipt of the data sets by NHS Digital, if further information is required in addition to that provided from the HES/Mortality data sites, the IBIS-I CCO will request supporting information from the participant's general practitioner (GP) or IBIS-I study file. The supporting information requested by the IBIS-I CCO would comprise cancer recurrence (i.e. grade, site, receptor status) and side effects (i.e. fractures, cardiovascular events, thrombo-embolic events). The s251 support covers the linkage to the GP data. Additionally, the IBIS-I Informed Consent Form (version dated 26/10/2001), approved by the REC, informed the participant that the above-mentioned further details will be requested from their GP. Processed data sets with all identifiers removed except the Study ID will be retained for analysis until study completion, following which it will be archived for 20 years. The original 'raw' data set returned by NHS Digital containing identifiable data will be stored on a separate server meeting NHS Digital security requirements, as per DARS Guidance Notes on Security for where Mortality data is involved. The applicant would like to emphasise that data is stored on the secure network and never on the local drives of unencrypted QMUL desktop PCs. QMUL creates back-ups of the data stored on its network. The encrypted backup tapes are stored offsite with Iron Mountain UK Ltd. The data is not accessible by any Iron Mountain employee. All personal identifiable data received at the Centre for Cancer Prevention are stored electronically on a database, local to and administered by substantive employees of QMUL only. Personal identifiers of study participants are stored separately to the clinical data, with access strictly restricted to only QMUL substantive members of staff. Only these members of staff will have access to the personal identifiable data. Additionally, access to the personal identifiers on the Oracle database is controlled by separate username and password access and is controlled by the IT Department. The separate Mortality and HES servers are self-contained within the Centre for Cancer Prevention network within the QMUL network. They are firewalled from external connections and the rest of the network. All traffic through the checkpoint firewall is logged. Access to data will be from within the network using workstations that have a currently supported operating system which includes security patches. Users are not permitted to download the data from these workstations. No remote access (i.e. through a VPN/RDP connection) is permitted. No mobile devices will be used to access any data. The network is externally scanned on a regular basis. The 'raw' data sets will be stored until the data destruction date as per this agreement with NHS Digital. The storage architecture is compliant with the NHS Digital Data Sharing Framework Contract and once QMUL is issued with a Data Destruction notice, the data from all storage including backups can be securely and permanently removed within 14 days. Data can be securely wiped to NHS Digital standards (multi pass pattern wiped to at least HMG S5 Enhanced on site and if end of life, degaussed and physically destroyed). Under this agreement, it is not permitted so share the "raw" data with international partners. Only aggregated data with small numbers suppressed in line with the HES Analysis Guide is permitted to be shared.