NHS Digital Data Release Register - reformatted

King's College London

Project 1 — DARS-NIC-68229-Y5J6V

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

Sensitive: Non Sensitive

When: 2019/08 — 2019/08.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(2)(c)

Categories: Identifiable

Datasets:

  • Hospital Episode Statistics Outpatients
  • Hospital Episode Statistics Admitted Patient Care
  • Hospital Episode Statistics Accident and Emergency

Objectives:

King’s College London (KCL) and South London and Maudsley (SLaM) NHS Foundation Trust requires HES data for the purpose of a research study called The Cognitive Behavioural Therapy for Dissociative (Non-Epileptic) Seizures (CODES). About 12-20% of patients who attend neurology or specialist epilepsy clinics because of seizures do not in fact have epilepsy but instead have dissociative (non-epileptic) seizures (DS). A high percentage of people with dissociative seizures will have other psychological or psychiatric problems and may have other symptoms. It is generally thought that people with dissociative seizures will benefit from psychological treatments. However, studies on this have been small or have not compared the psychological therapy with the treatment people normally receive (there is no standardised medical care (SMC)). There is some evidence that cognitive behavioural therapy (CBT) may lead to a reduction in how often people have dissociative seizures. CBT is a widely accepted psychology therapy that focuses on the person's thoughts, emotions and behaviour, and considers the physical reactions and sensations that may occur in their body. KCL has previously developed a CBT package for people with dissociative seizures. In a relatively small previous study, people receiving CBT overall showed greater reduction in how often they had their seizures. KCL is now leading on a larger study across several different hospitals, to obtain more definite results. The data processed by KCL is processed under Articles 6(1)(e) and 9(2)(j) as necessary for scientific research that is in the public interest. This is in line with KCL’s charter as an institution carrying out research in the public interest. KCL and SLaM do not consider there to be any moral or ethical issues raised by the dissemination of the data. The processing of personal data was clearly explained to participants who provided written consent for the research team to have access to health related records including records held by the Information Centre, the General Register Officer and other health related databases. At the time of recruitment, NHS Digital was known as the NHS Information Centre. Therefore, KCL/SLaM feel that participants have given them permission to access their data for the purpose described in this Agreement. The CODES study is a large multi-site study being funded by the National Institute of Health Research Health Technology Assessment programme (NIHR HTA). The study was submitted in response to a call from the NIHR HTA in 2012 to create a study to examine a treatment for this condition and to examine the health care costs to individuals and the NHS. The study is ‘co-sponsored’ by South London and Maudsley NHS Foundation Trust (SLaM) but SLaM does not determine any purposes of the study. The project is managed by staff at KCL and all patient data remains vested in KCL when the project completes. SLaM is provided with a small amount of research capability funding to facilitate the research because it requires NHS sponsorship. The only roles SLaM has is to process the grant funding and also an individual employed at SLaM is a co-investigator in the trial. Individuals from SLaM, the University of Sheffield, the University of Edinburgh and Royal Edinburgh Hospital are named as co-investigators in the study protocol. These individuals are part of the trial management group meaning they are equally responsible for the grant and collectively form one of the decision-making bodies of the trial. However, within the clinical trial there are different levels of decision-making. The decision to process HES data from NHS Digital was taken solely by the Chief Investigator from KCL and the decisions about how that data is analysed are taken by the Professor of Health Economics, also of KCL. The co-investigators are not data controllers. As co-sponsors for the trial, KCL and SLaM are joint data controllers for the purpose of this Data Sharing Agreement. The trial has 3 significant centres where organisations have contributed materially to the trial but are not processing the data: The University of Edinburgh, NHS Lothian and the University of Sheffield. They were recruitment sites but had dedicated trial staff who managed recruitment at the many trusts around these institutions. 45 NHS trusts took part in recruiting participants, a smaller number of these, 19, were treatment centres as well. Recruitment began in July 2014 and continued until May 2018 with 368 total participants consenting to be randomised across the UK. Of these, 284 participants consented in England and were not withdrawn at the end of the study. The data requested is for these participants only. All participants in the trial have provided a significant amount of data about their condition, how it impacts their lives and the lives of those who care for them, how their general health is, how they feel about their emotions, if they avoid things because of their seizures and many other facets of their lives. They did this up to three times during the trial period (baseline, 6-month follow up and 12-month follow up). 30 participants, just under 10% of the 368 who were randomised, also provided qualitative data by participating in a semi-structured interview with one of the researchers exploring their treatment, medical history, and experiences on the trial. The data KCL is requesting from NHS Digital contributes to a small part of the outcomes listed in the protocol. One of the trial's outcome measures, per the protocol, is to examine and cost participants' service use and KCL originally intended that this would come only from participants. However, in addition to the self-report, the funder (NIHR HTA) advised including a more objective measure of service use and identified centrally-recorded data on hospital admissions, routine attendances and emergency department attendances as the best measure of this service use from a non self-report measure. Participants who consented to the study were randomised to one of the two treatments. Service use for 6 months prior to each participant’s randomization will be requested and used as a baseline and the data from the 6 months prior to the end of each participant’s follow up period will be also be requested. In effect this means 6 months of data ending on the day of randomisation, followed by a six-month gap during which no data is requested from NHS Digital, and then another 6 months of data ending on the day the participant finished in the study. The dates for most participants will differ depending upon what day they were randomised. The primary outcomes for the trial are all measured 1 year after randomisation representing the beginning of treatment. The data will provide an objective measure of service use, without relying on the participant's memory. Furthermore, participants who suffer from dissociative seizures tend to have costly and inappropriate medical use, especially before a diagnosis is made (Mellers, 2005). Therefore, the use of this data will allow KCL to see whether diagnosis and treatment of dissociative seizures reduces service use - a question very important to answer when the NHS services and resources are stretched. Furthermore, health economics analysis will be carried out to assess the cost of service use in KCL’s sample, and to see if this cost is reduced during the study. The same analyses will be run on both the self-reported data and the data from NHS Digital. The findings of both analyses will be compared. The data will not be compared at individual participant level. Ideally, the analyses of both datasets should show the same levels of service use. If so, having two corroborating sources will make the findings more robust. If there are differences, the findings will be presented objectively and KCL will attempt to account for the differences. For example, there may be data entry bias in the HES data from NHS Digital or memory bias in the self-reported data. For the credibility of the study, it is important to present any differences and explain what they were attributed to. The data will only be accessible to the CODES study team at King's College London. The CODES team involves the Trial Manager, and the Professor of Health Economics and Junior Health Economist. All have computers based at King's College London, and data will only be stored on these computers. No other organisation will have access to the data.

Expected Benefits:

The overall aim of the CODES study is to determine whether being given a diagnosis of dissociative seizures and receiving one of two types of treatment reduces participants’ seizures, as well as their psychological distress symptoms and healthcare service use. The information KCL are requesting is in line with the information KCL have asked of the participants already, but this measure comes from system wide usage and will help avoid any forgotten, mis-remembered, or unreported data to provide an accurate and real reflection of cost and service usage. This use of this data from NHS Digital will enable KCL to conclude whether or not certain treatments reduces healthcare service use in patients with dissociative seizures. If it does, in the long term it will mean less money is being spent by the NHS which is very important as the NHS is currently financially strained. The outputs are expected in late 2019-20. The outcome of the health economics findings will be included in these outputs. Policy makers, CCGs, and other materially interested parties will be invited to a conference about the results with any health economics outcomes being extremely important. These outcomes, if sufficiently robust, could form the basis of any recommended changes to the health care system around this diagnosis and the treatment of people who have it. For example a CCG might decide to fund CBT therapy treatments for the condition because it is associated with better outcomes for the patient and a reduction on unnecessary A&E visits. The most significant change to the health care system and to patients in general would be a recommended standard care pathway for patients with this diagnosis. These changes rely in part on the outcomes from HES about the service usage. These changes could be incorporated into the health care system within a few years.

Outputs:

Findings for KCL’s baseline data, which would include data from HES, will be submitted by the middle of 2019, and KCL’s final outcome paper would come in late 2019 in a journal such as the Lancet or Journal of the American Medical Association. KCL will also publish results in the HTA's own journal which will include HES data. KCL are committed, as is KCL’s funder, to open access journals so that this information is available as freely as possible. Findings from the study will be communicated back to CODES participants through a specially written document posted to them after KCL’s analysis is complete. KCL will also be including an alternative infographic version of the findings where possible to convey the information as succinctly and easily as possible to the participants. The findings will also be reported on the trial website (http://www.codestrial.org/) which contains information for patients (not limited to trial participants) and health professionals (see: http://www.codestrial.org/information-booklets/4579871164). As the CODES trial is a first trial of its kind and the largest ever done with people who have dissociative seizures, KCL’s work will be of interest to many conferences. KCL will submit presentations to the following conferences, which would include results from KCL’s analysis of the data provided by NHS Digital. British Neuropsychiatry Association, International League Against Epilepsy- British Chapter- annual conference, American Epilepsy Society annual conference, European Congress on Epileptology, International Congress on Epileptology, Annual Conference of the British Association for Behavioural & Cognitive Psychotherapies, and the Annual meeting of Association of British Neurologists. All outputs will be in aggregate form only with small numbers supressed in line with the HES analysis guide.

Processing:

KCL will submit the NHS number, study ID (Participant Identification Number or PIN), and two dates to NHS Digital for each participant. The dates for each participant are the date six months prior to randomisation and the date six months prior to completion of follow up for each individual. All participants have consented to this information being passed to NHS Digital for this purpose. The information KCL is requesting has also been requested from the participants through a self-report measure but receiving the HES data from NHS Digital will provide an independent measure. The information KCL receives from NHS Digital is not going to be directly compared to the individual’s self-report. Instead both self-report and the data provided by NHS Digital are summarized and compared on an aggregate level. Using the NHS Number, NHS Digital will extract and provide the HES data requested for each participant. The data requested relates to HES outpatient, admitted patient care and A&E visits only. The returned data will contain no directly identifying details but will include the participants unique study ID. The linked data will be provided by NHS Digital to the study team and downloaded by the Trial Manager onto the highly encrypted secure hosting environment provided through KCL’ s protected digital network storage. It will be stored, processed and linked only through the server hosted by KCL which can only be accessed by substantive employees of the university: the Trial Manager, Chief Investigator, Professor of Health Economics and the Junior Health Economist. All individuals with access to the data are substantive employees of King’s College London. All processing activities will take place within KCL. The data set will be kept and stored separately to the identifying details and will include the study ID number. The data will not be linked to other data held by KCL, with the exception of a treatment identifier indicating whether the participant received CBT and SMC or SMC Alone. These are the two treatments the trial is comparing. The data from HES will contribute to a health economics analysis which is interested in average costs and service use in patients. The data provided from HES will be compared at an individual level to see if participants’ service use (and the costs associated with this service use) changed during the study. This will contribute to an understanding of whether or not CBT and SMC or SMC alone are related to a decreased use of services, and if so how much savings that translates to taking into account the cost of the treatment. The HES data will not be directly compared to the self-report data on an individual level. Instead the aggregates of service use change within the HES data will be compared at aggregate level to the self-report data to see if they both show similar trends. All reports/outputs will be aggregated with small numbers suppressed in line with the HES analysis guide.


Project 2 — DARS-NIC-387635-C9Y0W

Opt outs honoured: Y, N, Yes - patient objections upheld (Mixture of confidential data flow(s) with consent and flow(s) with support under section 251 NHS Act 2006)

Sensitive: Non Sensitive, and Sensitive

When: 2016/04 (or before) — 2020/02.

Repeats: One-Off, Ongoing

Legal basis: Informed Patient consent to permit the receipt, processing and release of data by the HSCIC, Section 251 approval is in place for the flow of identifiable data, Section 42(4) of the Statistics and Registration Service Act (2007) as amended by section 287 of the Health and Social Care Act (2012), Health and Social Care Act 2012 – s261(7), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant, Identifiable

Datasets:

  • Hospital Episode Statistics Admitted Patient Care
  • MRIS - Cause of Death Report
  • MRIS - Scottish NHS / Registration
  • MRIS - Flagging Current Status Report
  • MRIS - Cohort Event Notification Report

Objectives:

ONS: The Royal College of Physicians (RCP) is the data processor responsible for producing the CCG Outcomes Indicator Set (CCGOIS) measure of mortality at 30 days for stroke patients. These results are provided to the HSCIC to publish as part of the wider CCGOIS. The results are also provided at team level to provide necessary context on the performance of clinical teams treating stroke patients. As well as reporting on 30 day mortality, there is a need to show survival at other intervals such as at 6 months and 1 year. The outputs of the analysis by RCP will include mortality statistics at different time points and at different levels of granularity and dates of death will be used in statistical modelling. Any data reported on is carefully considered in terms of whether it could be potentially identifiable and advice is given on how the outputs should be interpreted. It is also important that Royal College of Physicians are able to provide the information back to the clinical teams who have treated the patients. HES: The HES dataset is used to determine the case ascertainment (case ascertainment is a measure of the number of cases reported in the audit, compared to the number of cases identified in HES) of participants of the Sentinel Stroke National Audit Programme (SSNAP), that is, the proportion of coded stroke patients which are recorded in the audit; and identify any readmissions and further strokes, in order to compare quality of care with outcomes for patients. As the outputs of analysis of SSNAP are reported and publically available, the proportion of patients entered into the audit for each hospital team, compared with the numbers in HES, is vital in determining how results are used (for instance, if there is low case ascertainment, the mortality outcomes would not be reported so that there is no potential misrepresentation).

Yielded Benefits:

Case ascertainment information from HES has been used to target trusts who were not achieving good levels of data entry to the audit in previous years, which has resulted in those trusts entering more records onto SSNAP, therefore improving the overall case ascertainment of the audit and reducing potential biases. This results in higher quality data being used for decision making at trust level and nationally. Mortality information has been fed back to trusts in case-mix adjusted models, and outlier trusts have been identified. These trusts were contacted and encouraged to undertake case note reviews of their fatalities to identify areas for improvement. Outlying trusts were also offered a full peer review visit by the Stroke Programme, and a number of outlying trusts have taken up this offer to help identify where improvements in their service need to be made. The ability to adjust for variables such as stroke severity using the SSNAP Civil Registration/Mortality methodology is important, as stroke severity is a very strong predictor of mortality. Statistical analyses of the HES and Civil Registration/Mortality data have looked at variation in stroke care and outcomes based on the presence of other diagnoses, socioeconomic status, and organisational characteristics of the hospitals treating the patients. Some of these analyses have already been published, and others are currently being written up for submission to peer review journals. In 2018 the paper ‘Socioeconomic disparities in first stroke incidence, quality of care, and survival’ was published in The Lancet. It is available here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5887080/. Such analyses have the potential to highlight key areas for improvement and to drive change.

Expected Benefits:

Case ascertainment information will be used to target trusts who are not achieving good levels of audit case ascertainment, this leads to more complete data and more valid results in future audit. Complete audit information is essential for service improvement, and improvements to stroke patient care. Mortality within 30 days of hospital admission is part of Domain 1 of the NHS CCG Outcome indicator set – ‘reducing premature mortality’. CCGs will access the published information and use it to improve services through identification of good and bad practice. This will be of benefit both in terms of better value for money and better patient outcomes. Similarly, trusts will use team level mortality within 30 days of hospital admission to identify trends and good practice, again leading to better patient outcomes.

Outputs:

ONS and HES: Indicators will be produced showing the performance of organisations and at national level for the purpose of monitoring and quality improvement, in particular: • Mortality within 30 days of hospital admission for stroke CCG Outcomes Indictor Set (CCGOIS) at least annually (first publication on 17 December 2014 next publication anticipated to be published by the end of 2016) • Mortality within 30 days of hospital admission for stroke Team-level mortality results (published in line with CCGOIS and used for contextualising the results). (Team usually equates to a hospital). • Audit case ascertainment information • For statistical purposes such as monitoring trends registered individuals at Trusts can access date of death for patients they submit to the audit derived from ONS mortality data.

Processing:

RCP will send cohort information to the HSCIC for linkage, they send NHS Number, Full postcode, Name, and a unique SSNAP ID. As part of the section 251 support, there is a method by which the information is sent to HSCIC for linkage without the RCP viewing any patient identifiable information. The HSCIC return; • Non sensitive pseudonymised HES data with SSNAP ID for patients in cohort • Non sensitive pseudonymised HES data for patients with a diagnosis of stroke • Identifiable ONS date and cause of death RCP combine HES and ONS data with SSNAP data and combine into separate databases; one with SSNAP and ONS data and the other with SSNAP and HES data. Identifiers are held separately to other data and the pseudonym SSNAP ID is used to identify individual patients. With the exception of date of death, analysts access no identifiers. Pseudonymised HES Data is then analysed to calculate case ascertainment information for the audit. HES data is also used to validate some of the information collected in the audit. Identifiable ONS data is analysed to produce 30 day mortality at CCG level and stroke team level (team usually equates to a hospital). Cause of death is used to disaggregate stroke specific deaths and deaths from other causes. For statistical purposes such as monitoring trends identifiable ONS death data is also passed back to registered individuals at participating trusts whereby they can access date of death for patients they submit to the audit. All arrangements for 3rd party access will be controlled through sublicensing agreements and will be for the benefit of health and care; all arrangements will be approved by the HSCIC before data being sent. All individuals with access to the data are employees of the data processors detailed in the application


Project 3 — DARS-NIC-309328-R9V1C

Opt outs honoured: Y

Sensitive: Non Sensitive

When: 2016/04 (or before) — 2016/08.

Repeats: Ongoing

Legal basis: Section 251 approval is in place for the flow of identifiable data

Categories: Identifiable

Datasets:

  • MRIS - List Cleaning Report

Objectives:

Data will allow King’s College London (KCL) to locate patients and invite them to participate in a study called STRATA. The patients were originally involved in a study called AESOP and were recruited into the study in 1998-1999 at one of two sites (Nottingham and South East London). AESOP recruited first episode psychosis studies to investigate the epidemiological factors relating to first episode psychosis in a cohort representative of the local population. Informed consent was taken for AESOP, and while it was not standard practice at that time to take consent for re-contact, the Section 251 approval permits KCL to contact patients who did not explicitly give consent to be contacted about future studies. The objective of STRATA is to explore the predictors of treatment-resistant schizophrenia. Previous studies examining treatment-resistance have recruited patients with chronic schizophrenia and have therefore been biased towards more treatment-resistant patients. STRATA includes only first episode patients to minimise this bias. STRATA will therefore have proportions of treatment-resistant and treatment-responsive patients who reflect those found in the general population of schizophrenia patients. However, as it can take an average of four years to determine whether a patient is treatment-resistant, STRATA is including first episode studies where the patients have been followed up for some time. The most recent AESOP follow up time point was 10 years. Some patients were unable to be traced and staff issues at the time meant that blood samples were not collected from all the patients who were traced. Since a key hypothesis of STRATA is that genetic factors will be able to predict which patients are resistant to antipsychotic medication and which will respond to antipsychotic, all patients included in the study must have DNA data extracted from a blood sample.

Expected Benefits:

While the use of list cleaning data will not have any immediate benefits to health and/or social care, the data obtained from patients who consent to take part in the study could lead to considerable advancements in the care of patients with schizophrenia within 5-10 years. Identifying a stratifier of treatment resistance in schizophrenia could allow clinicians to identify medication more suited to their patients, far more quickly than current clinical practice. This could save some patients many years in hospital on medication which has considerable side effects and is not the best medication strategy for them.

Outputs:

If patients, contacted using this data, consent to participate in the study they will provide a blood sample and history of their medication use. This data will be pseudonymised and combined with other data to form part of a wider analysis to examine treatment-resistant schizophrenia. Patient data will be connected to a unique ID number. This ID number will also be on patients’ consent form, which is the only place their name, address, and identifiable details will be stored. The data from these patients will be combined with data from other patients who have taken part in in multiple first episode studies from around the world. In this final dataset, totaling approximately 3000 patients, patients will only be identified by a unique ID number (please note that this will be different to the ID number used when ex-AESOP patients are seen). STRATA will receive no identifiable patient information (names, addresses) other than date of birth. This dataset is the combined output which will be shared with other researchers working on STRATA. It will also be combined with a separate part of STRATA, which is approved under a separate ethics application, which is recruiting patients with chronic schizophrenia for an imaging study. Combined, the imaging study results and results from the sample of 300 first episode patients who have been followed up will be used to develop a stratifier of treatment response. This stratifier will be a method (imaging, genetics, or a combination of the two) which can be used in clinical practice when a patient first presents with psychotic symptoms to determine whether they will respond to conventional antipsychotic medication or not. STRATA is planning a clinical trial, which will be pre-registered and is due to start in 2017, to test this stratifier in a clinical setting. Other parts of STRATA are working on the socioeconomics of using such a stratifier and patient’s opinions of the acceptability of such a stratifier (the latter of which has been published; Service user's and carer's views on research towards stratified medicine in psychiatry: A qualitative study. Rose, D., Papoulias, C., MacCabe, J. & Walke, J. 28 Sep 2015 In : BMC Research Notes. 8, 1, 489). The results produced by STRATA will be published in peer review journals (for example, Biological Psychiatry, Schizophrenia Research, etc.) and presented at conferences (for example, the Schizophrenia International Research Society Conference, the World Congress of Psychiatric Genetics, etc.)

Processing:

The details on patients who were originally recruited at the Nottingham site will be sent via secure transfer to a specified user at Nottingham University. The details on patients who were originally recruited in South London will be retained by a specific user at KCL. Addresses or GP addresses will be used to send letters to patients. Or, if phone numbers are available, patients will be contacted by telephone using the telephone script which has been approved by the ethics committee. In all cases, the researchers will attempt to consult with the patient’s responsible clinician or care coordinator about whether the patient is well enough to be contacted. The study will be explained to the patients over the phone and they will then be given at least 24 hours to think about participating and discuss the study with family and friends. The patients will then be seen by the specified user at either Nottingham University or KCL and the study will be explained again, informed consent will be taken, a blood sample will be taken, and the patients will be asked to fill in a few questionnaires. This informed consent includes consent for re-contact regarding any future studies and up to date contact details. Once a patient has been seen, the information received from the HSCIC will be destroyed. The data will not be stored, processed or be in any other way accessible to a third party. Co-investigators based at other organisations (as mentioned in the study protocol) will only have access to aggregated outputs.


Project 4 — DARS-NIC-274251-H0G6M

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

Sensitive: Non Sensitive

When: 2019/12 — 2019/12.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant

Datasets:

  • Hospital Episode Statistics Admitted Patient Care

Objectives:

King’s College London (KCL) requires HES data from NHS Digital for the purpose of research which will use the National Hip Fracture Database (NHFD) linked to Hospital Episode Statistics (HES) and the Patient Episode Database for Wales (PEDW) to determine whether poor outcomes after hip fracture surgery were less frequent among patients exposed to more frequent, longer duration, and more comprehensive rehabilitation controlling for characteristics of patients, their injuries and healthcare. More specifically, the aims are to: 1. estimate the odds of return to preadmission residence, readmission and survival at 30-days post-discharge, and recovery of mobility at 120-days, overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders; 2. estimate the probability of discharge by time after surgery overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders and the competing risk of death; 3. determine whether estimates vary across subgroups defined by patient and care factors; The purpose and aims are justified under Article 6 (1) (e) and Article 9 (2) (i). This project will address an evidence gap with a view to complete analysis performed in the public interest of informing quality improvement in rehabilitative care for patients with hip fracture (Article 6 (1) (e)). This project will also help to reduce unwarranted variation in current rehabilitative care to ensure high standards of quality and safety of rehabilitative health care (Article 9 (2) (i)). KCL has determined that there are no moral or ethical issues raised by the proposed dissemination and there is no risk of potential harm to the public by the dissemination. UK hospitals admit 75,000 men and women with hip fracture annually. Patients with hip fracture and their carers describe rehabilitation as key to their recovery. Yet, the optimal rehabilitation remains unclear. This is highlighted by: the detail in the NICE guidance being limited to daily mobilisation and regular physiotherapy review; the absence of recent Cochrane systematic reviews; the conclusion that there is insufficient evidence to recommend practice change from earlier Cochrane reviews, and uncertainty among physiotherapists with respect to the appropriate management of patient subgroups. The National Hip Fracture Database (NHFD) is a clinically led, web-based quality improvement initiative commissioned by the Healthcare Quality Improvement Partnership (HQIP) and managed by the Royal College of Physicians (RCP). The NHFD audit demonstrated marked national variation in the duration, frequency, and type of acute rehabilitation delivered by physiotherapists during a fixed period in 2017. The audit recommended longer, more frequent, and more comprehensive rehabilitation. The audit was not resourced to evaluate the association between duration, frequency, and type of rehabilitation with outcomes, or possible variation in associations for different patient subgroups. The requested data will provide essential outcome data (readmission and survival at 30-days) as well as rich information on ethnic category, deprivation, comorbidities and complications for regression adjustment and subgroups in the proposed analyses. The aims will inform quality improvement initiatives to reduce unwarranted variation in rehabilitation after hip fracture. The data subjects will consist of all patients 60 years of age or older who underwent hip fracture surgery in England or Wales between May 1st and June 30th 2017. KCL requires HES Admitted Patient Care data for the years 2015/16 to 2017/18 linked to this cohort. KCL also requires the survival status at 30 days post discharge from care. KCL require pseudonymised data to address the aims outlined above which allows ‘the processing of personal data in such a way that the data can no longer be attributed to a specific data subject without the use of additional information’ (GDPR 2018). The additional information required to identify an individual will not be available at King’s College London. The NHFD collected data related to rehabilitation duration, frequency, and type for patients admitted with hip fracture between May 1st and June 30th, 2017. KCL require data for 2017/18 to capture the care spell related to hip fracture and any readmission within 30-days of discharge from the care spell. KCL also require data for care spells in the years prior to the hip fracture care spell (2015/16 & 2016/17) to identify comorbidities not coded in the hip fracture care spell for regression adjustment. KCL require data for England and Wales. KCL require this geographical spread to capture patients who underwent surgery at a NHFD site participating in data collection related to rehabilitation duration, frequency, and type which is required to address our study aims. KCL has determined that there are no alternative, less intrusive ways of achieving the purpose. The data is minimised to the cohort of patients whose care spell related to hip fracture between May 1st and June 30th, 2017. Care spells in the year prior to the care spell related to hip fracture, and Readmission within 30 days of discharge from the care spell related to the hip fracture. Survival status at 30-days derived from the civil registration mortality dataset will be the only mortality information requested. Kings College London is the data controller and also processes the data for this study. No other organisations process the data for this purpose. NHS Wales Informatic Service (NWIS) is a public service organisation which manages requests for access to administrative data including the PEDW. KCL will request similar data fields from NWIS as from NHS Digital to enable similar data fields for the analysis for patients in both England and Wales. The NHFD is a clinically led, web-based quality improvement initiative commissioned by the HQIP and managed by the Royal College of Physicians (RCP). Crown Informatics are RCP's data processor who will be sending in the cohort to both NHS Digital and NWIS for linkage. The Healthcare Quality Improvement Partnership (HQIP) is a joint Data Controller for the NHFD data but not for the HES data being released by NHS Digital or PEDW data being released by NWIS. RCP and HQIP will have no access to the data and will play no part in the processing of the data for this study. Researchers at the University of Oxford, University Hospital of Wales and University of Bristol will contribute to the interpretation of results, and preparation of materials for dissemination. They will have no access to the data and will play no part in the processing of the data for this study. This research is funded by the Chartered Society of Physiotherapy Charitable Trust. The funder has no role in the design of this study, execution, analyses, data interpretation, or overall dissemination plan. The applicants are required to present the results of the study at the annual Chartered Society of Physiotherapy conference as a condition of the funding award.

Expected Benefits:

This project will help to inform a reduction in unwarranted variation in current rehabilitative care to ensure high standards of quality and safety of rehabilitative health care and in turn promote patient health. This will be achieved by addressing the aims outlined above which will identify and report health inequities in access and delivery of rehabilitation after hip fracture and the impact of those inequities on outcomes, and dissemination of the findings to stakeholders involved in the receipt, delivery, and organisation of rehabilitation after hip fracture, as outlined in the outputs section. 75,000 people incur hip fracture each year. On average, these patients spend 15.5 days in hospital at £400 per day. If a reduction in unwarranted variation in rehabilitation duration, frequency, or type led to one less day in hospital for half of all admissions this would reflect a cost saving of £15,000,000 each year. It is therefore in the public interest to ensure high standards of quality and safety of rehabilitative health care for these people. It is appropriate for the proposed analysis to be completed and results disseminated as they will inform quality improvement initiatives to reduce unwarranted variation and improve the standards of quality and safety of rehabilitative health care. In 2017, the National Hip Fracture Database audit demonstrated marked national variation in the duration, frequency, and type of acute rehabilitation delivered by physiotherapists. The audit was not resourced to evaluate the association between duration, frequency, and type of rehabilitation with outcomes. The outputs will address all aims outlined above and therefore determine the association between duration, frequency, and type of rehabilitation and outcomes. Moreover, the outputs will be disseminated to all stakeholders to facilitate implementation of quality improvement initiatives to reduce unwarranted variation. The requested HES data contributes a significant impact to the aims outlined above. First, it will enable KCL to determine the association between rehabilitation duration, frequency and type and readmission and survival at 30-days. These outcomes are key to informing future quality improvement initiatives. Second, without these data, the analyses will fail to control for potential confounders for the putative association between rehabilitation duration, frequency, and type on outcomes. Third, without these data, KCL will not be able to determine whether different subgroups of patients respond differently to rehabilitation duration, frequency, and type. These analyses are likely to inform a reduction in unwarranted variation in access and delivery of rehabilitation to improve outcomes by identifying and reporting health inequities (as they relate to outcomes). Patients and the NHS will achieve the benefit through improvements in standards of quality and safety of rehabilitative health care. The benefit will be measured through successful completion of the dissemination strategy outlined in ‘Specific outputs expected, including target date’. The analysis will be completed and dissemination 24 months after release of data.

Outputs:

An interim report on the progress of the research will be sent to the funding body (Chartered Society of Physiotherapy Charitable Trust) in June 2020. A final report to the Chartered Society of Physiotherapy Charitable Trust will follow in June 2021. This will cover results to address all aims outlined above. Planned academic papers submitted to open-access peer reviewed journals: 1 - the odds of return to preadmission residence, and recovery of mobility at 120-days, overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders (December 2019); 2 - the odds of readmission and survival at 30-days post-discharge overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders (January 2020); 3 - the probability of discharge by time after surgery overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders and the competing risk of death (May 2020); 4 - variation in the odds of return to preadmission residence, and recovery of mobility at 120-days, overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders across subgroups defined by patient and care factors (August 2020); 5 - variation in the odds of readmission and survival at 30-days post-discharge overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders across subgroups defined by patient and care factors (December 2020); 6 - variation in the probability of discharge by time after surgery overall and by duration, frequency, and type of rehabilitation after adjustment for potential confounders and the competing risk of death across subgroups defined by patient and care factors (May 2021); A lay version of the findings of each paper will be disseminated to patients and the public interested in research on hip fractures through the Royal National Osteoporosis Society. For each paper published, briefing papers and PowerPoint slide decks will be developed to summarise the findings for a range of stakeholders including healthcare professionals, policymakers, patients and their caregivers. Findings will be presented at the Fragility Fracture Network 2020 and 2021, the British Geriatric Society 2020 and 2021, as well as the Chartered Society of Physiotherapy 2021. All publications and conference presentations will be promoted on twitter via the Kings School of Population Health and Environmental Sciences account (>1000 followers), the Falls & Fragility Fracture Audit Programme account (>1500 followers) and study personnel (>2000 followers). All outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide. KCL will publish outputs in peer-reviewed open-access journal articles and in reports for the Chartered Society of Physiotherapy Charitable Trust. In the UK KCL will work with the Chartered Society of Physiotherapy to disseminate findings to the national community of physiotherapists through their website and publication Frontline. KCL will communicate findings nationally through the FFFAP, the British Orthopaedic Society, the British Geriatrics Society, and the Royal College of Occupational Therapists to substantiate their position on access to therapy. KCL will present findings locally (London, Bristol, Bath, Cardiff, Oxford, Norwich hospitals), nationally (Physiotherapy UK, British Geriatrics Society and British Orthopaedic Society conferences), and internationally (Fragility Fracture Network conference). KCL will continue to partner with the National Osteoporosis Society to disseminate findings to the public via their various media streams: updates/articles in their quarterly membership magazine Osteoporosis News (23,000 readership); their Bone Matters e-newsletter (21,000 readership); and on the research section of the website (www.nos.org.uk/research). KCL will provide briefing papers and PowerPoint slide decks to policymakers who influence hip fracture quality care indicators for Best Practice Tariffs such as Public Health England, FFFAP, and NHS Improvement. KCL will also present briefing papers and slide decks to NICE in preparation for future updates to the NICE Hip Fracture guideline (CG 124). Internationally, KCL will work with the Fragility Fracture Network physiotherapy and rehabilitation special interest groups. KCL will disseminate findings to the International Coordinating Council for the Bone and Joint Decade, the International Osteoporosis Foundation, and the European Geriatric Medicine Society. All publications and conference presentations will be promoted on twitter via the Kings School of Population Health and Environmental Sciences account (>1000 followers), the Falls & Fragility Fracture Audit Programme account (>1500 followers) and study personnel (>2000 followers). The dissemination of research outputs outlined above includes communication activities to target all stakeholders including patients and the public, clinicians, researchers, and policymakers. KCL will also work with the School of Population Health and Environmental Sciences communication team to develop press releases for media engagement.

Processing:

Crown Informatics (RCP’s data processor) will send the following identifying details for the cohort to NHS Digital: - Study ID - NHS number - Date of Birth - Surname - Forename - Postcode - Date of Admission NHS Digital will receive the cohort identifiers detailed above. NHS Digital will identify the relevant care spells for patients in HES data and will extract the data for the care spell related to hip fracture and any readmission within 30-days of discharge from the care spell. They will also extract the same data for care spells in the year prior to the hip fracture care spell. Fact of Death sourced from the Civil Registration Mortality dataset will also be extracted for the cohort and will only be used to indicate the survival status at 30 days following discharge from the care spell. NHS Digital will send the following pseudonymised fields to King’s College London: - Study ID - [ADMIDATE] Date of admission, - [ADMISORC] Source of admission, - [DIAG_NN] All Diagnosis codes, - [DISDEST] Destination on discharge, - [ETHNOS] Ethnic category, - [IMD04 IMD] Index of Multiple Deprivation, - [IMD04HD IMD] Health and Disability Domain, - [IMD04RK] IMD Overall Rank, - [STUDY_ID] STUDY_ID - [DISDATE] Date of discharge - Survival status at 30-days (Calculated from Date of Discharge) - Indicator of readmission within 30-days There are no subsequent flows of data. The flow of data from Crown Informatics into NHS Digital will include personal data with identifying details. The flow of data from NHS Digital to KCL will not contain identifying details. Kings College London will process data from NHS Digital, as well as data received from NWIS and Crown Informatics. This data is personal data and will not contain identifying details. Crown Informatics is processing data that will flow to NHS Digital, NWIS, and Kings College London. Crown Informatics will identify the study cohort as all patients 60 years of age or older who underwent hip fracture surgery in England or Wales between May 1st and June 30th, 2017. Crown Informatics will then extract the cohort data detailed in this purpose statement and send to NHS Digital for England, and NWIS for Wales. Crown Informatics will then remove the identifying data except for the Study ID. Crown Informatics will also minimise the audit data requested to variables relevant to the stated purpose. Crown Informatics will send pseudonymised audit data including the Study ID to KCL. NWIS will also receive the cohort details from Crown Informatics. NWIS will identify the relevant care spells for patients in PEDW data. NWIS will extract the data for the care spell related to hip fracture and any readmission within 30-days of discharge from the care spell, including survival status at 30 days. They will also extract the same data for care spells in the year prior to the hip fracture care spell. NWIS will send pseudonymised audit data including the Study ID to KCL. KCL will link the data received from NHS Digital and NWIS to the data received from Crown Informatics using the Study ID. KCL will then process the pseudonymised data to meet the purpose and aims outlined above while enabling ‘the processing of personal data in such a way that the data can no longer be attributed to a specific data subject without the use of additional information’ (GDPR 2018). The additional information required to identify an individual will not be available at KCL. There will be no requirement/attempt to re-identify individuals. The data processing at KCL will only be carried out by substantive employees of KCL who have completed KCL’s GDPR training and NHS Level 1 Data Security Awareness Training and who have confirmed in writing they have read and understood the protocols outlined and referred to in the Data Security and Protection Toolkit. The data will be held as an encrypted file on KCL’s secure network drive. The encrypted file will be held in an access-controlled area of the network drive and the data will be accessible to authorised study personnel only, by means of a password or a recovery key. All users have their own username and password, and these will never be shared. Access to the data for the study will be cancelled as soon as a user leaves the study, KCL, or if they are absent for a long period. Data will be stored on servers owned and managed by KCL, within the JISC Southern Data Centre. Only individuals with the administrative rights to those servers will be able to access data stored on them. The only individuals with those administrative rights are KCL’s IT Computer & Storage team. JISC is a United Kingdom not-for-profit company whose role is to support post-16 and higher education, and research, by providing relevant and useful advice, digital resources and network and technology services, while researching and developing new technologies and ways of working.


Project 5 — DARS-NIC-199726-F4V3C

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

Sensitive: Sensitive, and Non Sensitive

When: 2019/09 — 2019/09.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant

Datasets:

  • Civil Registration - Deaths
  • Hospital Episode Statistics Outpatients
  • HES:Civil Registration (Deaths) bridge
  • Hospital Episode Statistics Admitted Patient Care

Objectives:

Aim and purpose of the application: The Healthcare Quality Improvement Partnership (HQIP) are the sole data controllers. HQIP commissioned the British Society for Rheumatology (BSR) to undertake the NEIAA as part of the National Clinical Audit and Patient Outcomes Programme (NCAPOP). The British Society for Rheumatology is the UK's leading specialist medical society for rheumatology and musculoskeletal professionals. The BSR subcontracted an academic unit at King's College Hospital NHS Foundation Trust (KCH) to carry out the data processing, including all analyses and linkage. The Healthcare Quality Improvement Partnership (HQIP) requires hospital episodes statistics (HES) and mortality data for use in the National Early Inflammatory Arthritis Audit (NEIAA). This audit will help to improve the quality of care for people living with inflammatory arthritis across England and Wales. The current contract period is 1 October 2017 – 30 September 2020, with a further planned two year extension. The aim is to improve the quality of care for people living with inflammatory arthritis by assessing the performance of rheumatology units against NICE Quality Standards. There is compelling evidence that early intensive treatment greatly improves the outcome of these disabling diseases, which predominantly affect people of working age. Early diagnosis and treatment is a cornerstone of Early Inflammatory Arthritis (EIA) management and is underpinned by NICE guidelines (CG79). The audit will assess EIA services and will collect prospective data including: • Waiting times; • Time to treatment; • Provision of education; • Collection of patient reported outcomes; • Clinical response. • What’s included: – NHS secondary care settings in England and Wales. • What’s excluded: – Children and children’s services – Primary care The linkage requested is necessary for the performance of a task carried out in the public interest; improving the quality of care for people living with inflammatory arthritis (covered by Article 6 (1)(e) of GDPR). Processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject. The data requested will help to achieve the aim identified in the following ways: (1) Deliver quality metrics to inform Care Quality Commission (CQC) regulation of providers. (2) Create a dataset to inform quality improvement activity. (3) Create a dataset for epidemiological and health service research. (4) Quantification of the burden of disease for patients and society. (5) Provide evidence for cost effective service delivery. (6) Provide aggregate department level performance data for the Getting It Right First Time (GIRFT) programme Data subjects: Data are collected from all patients in England and Wales over the age of 16 who are seen in rheumatology services with a suspected diagnosis of early inflammatory arthritis. Purpose of Request: The collected audit data will be linked with the HES Outpatient and HES Admitted Patient Care dataset. It will also be linked the patient episode database for Wales (PEDW), pending approval from the NHS Wales Informatics Service. This will enable ascertainment of joint replacements, unplanned hospitalisations, and death. These linkages will be repeated annually. Only pseudonymised data will be requested - identifiers will be removed and a study ID will replace the identifiers. There will also be linkage to the Civil Registration/Mortality data set to determine mortality outcomes. In addition data on the total number of patients diagnosed with rheumatoid arthritis in outpatients for each trust will be requested. Data linkage will enable estimation of variation in the following: (1) Treatment delay (outpatient referral dates, diagnostic imaging dates), (2) Clinical outcomes (adverse events/unplanned hospitalisation, joint replacement surgery, mortality), (3) Healthcare resource utilisation (outpatient activity in 12 months following diagnosis), (4) Case-mix adjustment, and (5) Allow an assessment of case ascertainment. All data requested from NHS Digital will be pseudonymised data, this will allow patient level linkage while maintaining patient confidentiality. Linkage is requested for the duration of the contract of the project, and for all Trusts in England as NEIAA is a national project. There are no alternative less intrusive methods to achieve the above purpose of linkage. Data linkage will be reviewed on an annual basis to assess if the degree of data requested can be minimised. The data processing under this agreement is not in support of a specific PhD/post graduate research study, but may be utilised for future work, in the future. An amendment to this agreement or separate data sharing agreement will be formulated and submitted to NHS Digital for approval for this in the future if necessary.

Expected Benefits:

Dissemination of the NEIAA results will identify variation in early inflammatory arthritis care across England and Wales. Publication of this will reduce national variation in the quality of care in early inflammatory arthritis. More patients will attain a state of disease remission during their first year of treatment, fewer people leave work as a result of their arthritis, and overall quality of life for people diagnosed with inflammatory arthritis will improve. In order for the benefits to be achieved, outputs will be used by the CQC to identify outlier rheumatology departments. This will lead to increased scrutiny and support for under-performing departments. The NEIAA team have already engaged with the CQC, who have confirmed they will be utilising performance measures from the project to assess departments. In addition, aggregated data with small number suppression will be provided to the Getting It Right First Time Programme (GIRFT) to assist departments in improving the quality of care delivered. The Benefits: A central tenet of undertaking a National Audit is to deliver change that will extend beyond the local Trust level. (a) Deliver quality metrics to inform Care Quality Commission (CQC) regulation of providers. (b) Create a dataset to inform quality improvement activity. (c) Create a dataset for epidemiological and health service research. (d) Quantification of the burden of disease for patients and society. (e) Provide evidence for cost effective service delivery.

Outputs:

The following outputs will be produced: a. A publicly available report with site, Trust, clinical commissioning group (CCG), and regional level data, will first be published in July 2019, and will be repeated on an annual basis. The report will include performance against the NICE quality standards for early inflammatory arthritis. The report will only contain aggregated data with small number suppression applied in line with the HES analysis guide. b. Academic papers will be published in Rheumatology Journal on methodology, care variation, and impact of timely treatment on mortality and inpatient admissions. The BSR website will provide links to open access papers. The papers will only contain aggregated data with small number suppression applied in line with the HES analysis guide. c. For each paper published, a short presentation is developed to summarise the findings for a range of stakeholders, including healthcare professionals and patient groups. Findings will be presented at project working group meetings. d. Findings will be submitted for presentation at Rheumatology, EULAR, and ACR conferences in 2020. e. The website dashboard will provide case-mix adjusted departmental data as a result of the linkages obtained. As above, all data will be presented at aggregate level, with suppression of small numbers in line with the HES analysis guidance. Dissemination of results/outputs: Aggregated findings will be primarily disseminated in a publicly available annual report. Academic papers will be disseminated via peer reviewed journals. Key findings will be disseminated to rheumatologists via the BSR newsletter. All reports and open access journal articles with be accessible via the BSR website. Webinars providing updates on the audit are regularly made available on the BSR website. Communication of results/outputs: Summary level findings for health professionals and the general public will be available via the BSR website. Key findings will be reported publicly via social media. Again all results and findings will only contain aggregated data with small number suppression in line with the HES analysis guide. Exploitation of results/outputs. The data and knowledge collected in NEIAA are owned by HQIP, and managed by the BSR. Only aggregate data in annual reports will be open access. The first annual report is expected to published in July 2019.

Processing:

All organisations party to this agreement will 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). 1) Data flow to NHS Digital: KCH will share patient identifiable information including unique study ID, NHS number, date of birth, and postcode for the purpose of linkage to NHS Digital. NHS Digital will link these identifiers to the HES OP, HES APC and Civil Registration/Mortality Database. 2) Data flow from NHS Digital to KCH: NHS Digital will share the requested products (HES OP, HES APC, Civil Registration Data) alongside the unique study ID number provided by KCH. NHS Digital will share no identifiable information in the disseminated extract. 3) Data flow includes patient level information from providers to KCH and also from NHS Digital to KCH. Onward data flow from KCH to HQIP, the BSR and subsequent dissemination is aggregate, with redaction of data from sites with 5 or fewer subjects recruited to maintain anonymity (ie small number suppression in line with the HES analysis guide) 4) Data processing of the audit data (not NHS Digital data) takes place by Net Solving at the point of data entry from the providers. KCH then process the data for analysis, including preparing the linkage file for NHS Digital and then subsequently receiving the NHS Digital linked data. 5) Net Solving manage the online data entry portal and the data extract too for the audit data. Net Solving undertake no data manipulation and will have no access to the data disseminated by NHS Digital. KCH then process the data for the purpose of answering the specific questions set out by HQIP for NEIAA. KCH also receive the NHS Digital linked data for the same purpose. KCH produce aggregate data reports using the NHS Digital data. 6) The online portal entered data are linked to IMD rank (the Index of Multiple Deprivation (IMD) is a measure of relative deprivation for small areas), using postcode. The clinical data-set will be linked to the disseminated NHS Digital data extract using the IMD rank not postcode. 7) IMD rank is publicly available, and will be linked via the patients' postcode. After which the postcodes will be deleted from the data-set to reduce the risk of identification. 8) Linked data will be associated with a unique study ID. All patient identifiable data will be removed to prevent re-identification once data has been linked. 9) There will be no requirement/attempt to re-identify individuals within the extract. 10) Data provided by NHS Digital will only be accessible by individuals within the academic team at KCH who have authorisation from the academic lead to access the data for the purposes described. All individuals accessing data will have undergone GDPR training, and are substantive employees of the named data processors on the agreement. 11) Access to the NHS Digital data will only be granted to authorised substantive employees of the data processor. Third party organisations would have to make a formal NHS digital data access request in order to obtain the data-set. 12) KCH will store the data on a KCH secure server, which can only be accessed on site. Once linkage has occurred, the patient identifiable data used for linkage held by KCH will be destroyed. Linked data will never be stored in the same location as participant identifiable information. 13) The data will be stored on the premises of the data processor. 14) Summary level reports will be provided to HQIP and the BSR for public use and dissemination from KCH after the data processing has taken place. These summary level reports will only contain data that has been aggregated with small number suppression in line with the HES analysis guide. 15) All aggregated reports will suppress small numbers in line with the HES analysis guide. There will no linkage permitted to other data sets apart from what is detailed in this agreement. There will be no attempts by employees of the named data processors or controllers to re-identify participants in the audit.


Project 6 — DARS-NIC-174740-C0H0L

Opt outs honoured: No - data flow is not identifiable (Does not include the flow of confidential data)

Sensitive: Non Sensitive

When: 2019/07 — 2019/07.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant

Datasets:

  • Hospital Episode Statistics Admitted Patient Care

Objectives:

South London and Maudsley NHS Trust and King’s College London require HES Admitted Patient Care (APC) data for use in a project entitled ‘Understanding variations in self-harm rates between deprived areas in London’. This study forms the final part of a PhD project examining variations in the rates of self-harm between small-areas in London. The applicant is employed by King’s College London (KCL) and registered as a PhD student within the Department of Psychological Medicine at KCL. They also do clinical work as a psychiatrist for South London and Maudsley NHS Foundation Trust (SLaM) and has a clinical addendum to their contract with KCL which establishes an honorary contract for this work. The National Institute for Health Research (NIHR) Maudsley Biomedical Research Centre (BRC) is part of SLaM and works in partnership with the Institute of Psychiatry, Psychology & Neuroscience at King's College London. Together they aim to develop more individualised treatments and support advances in the prevention, diagnosis, treatment and care of mental ill health and dementia. To do this, they bring together researchers, clinicians, allied health professionals and service users from across the University/Trust partnership to work together better in order to meet the challenges of finding better treatments and improved care for patients. Staff working within the BRC are SLaM employees, all data for BRC projects is held on the SLaM network and owned by SLaM and individuals employed outside SLaM who access BRC resources have to hold honorary contracts with SLaM. The earlier studies within the PhD project have used non-NHS Digital data held by SLaM. The applicant has worked with this data within the SLaM network. Within the Maudsley BRC, SLaM has expertise, security and information governance frameworks in place for hosting clinical data. For this reason, it is proposed that the data requested for this study will also be held by SLaM. No data will be stored by KCL and it will not have access to any record level data supplied by NHS Digital. Hence, SLaM will be the sole data processor. The study was devised by the applicant as part of the work towards the PhD they are completing at KCL. The PhD is funded through a clinical research training fellowship awarded to the applicant by the Wellcome Trust. This study forms the final part of a PhD project examining variations in the rates of self-harm between small-areas in London. Work prior to this study is currently using a clinical dataset already held by the Maudsley BRC, of non-NHS Digital data, covering an area of South East London. This work is testing associations between area exposures and rates of hospital admission for self-harm for people aged 11 and over living in four London boroughs. Subsequent work will build a model that aims to predict rates of self-harm based on area-level variables. The study this application relates to aims to 1) Describe the distribution of self-harm hospital admission rates across small-areas of Greater London and over time and compare them to the distribution of all admissions to identify any self-harm specific spatial patterning. 2) Explore the impact of using different definitions of self-harm, for exampling including injuries and poisonings coded as of undetermined intent or accidental, on associations with self-harm rates and the geographical patterning of self-harm rates. 3) Test the validity of a model using area and population level exposures to predict areas with high and low rates of self-harm. Data for this study has not been provided by NHS Digital before. The data required from NHS Digital is pseudonymised Hospital Episode Statistics Admitted Patient Care data. This will be used to identify the outcome of interest for the study: admission to hospital for self-harm. The NHS Digital data will not be linked with any other data sets and there will be no attempt to identify individuals. Data is required for individuals living in Greater London at the time of admission, for the years 2008-2017 as this is the study area and period of interest. Data for all individuals aged 11 or older is being requested so that the dataset the model is being tested on matches the clinical data that was used to build the model in this regard. Data for younger individuals is not being requested as incidents of self-harm are rare prior to adolescence. Data would be at admission level to allow the calculation of admission rates. Fields required include age in years and sex to allow standardisation of rates, lower super output area (LSOA) of residence to allow small area rates to be calculated, ethnicity, date of admission, hospital of admission and ICD-10 diagnostic codes.

Expected Benefits:

Self-harm results in over 100,000 admissions to hospital in England each year. For individuals, self-harm requiring medical attention represents mental distress and usually disorder, damage to physical health and is the strongest single risk factor for future suicide. For health care services it represents a substantial proportion of presentations for emergency care, especially among young people. At a population level, preventing and responding to self-harm and enhancing community based support have been identified as key priorities within Public Health England’s suicide prevention strategy. Local authorities’ public health departments are tasked with drawing up local suicide prevention plans based on these priorities. Targeting of preventative interventions and support services for self-harm at a local level could be enhanced by a better understanding of where areas of increased risk are likely to be. However, statistics on service use for self-harm are not routinely available at a small area level, and the statistical techniques required to reliably interpret spatial patterns across areas with relatively low counts are complex. Furthermore, area-level factors that are known to be predictive of self-harm rates nationally, especially deprivation, have been shown to be less strongly predictive of self-harm rates within London, making them less useful as a basis for targeting services. This project aims to validate a model to predict areas likely to have higher or lower than average rates of self-harm within Greater London. Such a model would provide benefits to local authority public health departments in informing their local suicide prevention plans and the targeting of preventative interventions. It could also help inform decision making by mental health service providers, community support organisations and commissioners in selecting the most appropriate locations for services working with individuals who self-harm. A model based on area characteristics available through publicly available data would enable such users to characterise the level of need in an area quickly and without having to set up local data collection systems or access more sensitive clinical data for an area. As outlined under the outcomes section, the PhD student aims to submit a paper describing the development and testing of the predictive model including the use of data being applied for to a peer reviewed journal by May 2020. The project has established contacts with the public health departments local to KCL and with local community groups working in mental health who are keen to make use of data such as this in their local suicide prevention plans. Further dissemination more broadly across London public health teams and mental health services is planned between March and September 2020, with support from local contacts and by using Public Health England London Mental Health Network, to ensure these benefits are realised.

Outputs:

The proposed project forms part of a larger PhD project examining the reasons for variations in the rates of self-harm between small areas in London, and area level factors that may be associated with them. It arose from a literature review that found that the rates and predictors of self-harm in London appear to differ from those in England as a whole. The project described above aims to test the validity of a model being developed by preceding work using data from South East London on data for all of Greater London. The model will use routinely available area-level data to predict which small areas within London have above or below average rates of self-harm. The validity testing described above will result in a model that can be applied to the whole of Greater London. Outputs from the project will include peer reviewed papers in academic journals. The project aims to submit the final paper of the PhD project, which will describe the development and testing of the predictive model including the use of data being applied for, by May 2020. Target journals for this paper are journals with a wide audience in general and public health, for example the British Medical Journal or Journal of Public Health. Any paper will be available as open-access. The work will also be described in the applicant's PhD thesis, which will be submitted in September 2020. Conference presentations to academic, policy and practitioner audiences within public and mental health are planned between November 2019 and September 2020. Target academic conferences include the European Public Health Conference and the Society of Social Medicine. In addition, the project plans to share findings with local mental health services, which work with individuals presenting with self-harm, and with public health departments, who write local suicide and self-harm prevention plans and commission mental health services, through seminars and presentations at team meetings and more widely across London through the Public Health England London Mental Health Network between March and September 2020. In addition, lay summaries of the research will be disseminated to a wider lay audience. This dissemination will take place in collaboration with community organisations with an interest in mental health, who the PhD student is working with during earlier work as part of the PhD project, and through the Health Inequalities Research Network (HERON) engagement events and website and through the Maudsley BRC’s engagement events and website. This work will take place throughout the PhD project, with results relating to the data requested in this proposal being disseminated between March and September 2020. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

Processing:

Data flows: The data requested is pseudonymised and will not be linked to any other data. Hence there will not need to be any data provided to NHS Digital. HES APC data will be provided by NHS Digital to SLaM. It will be stored within the SLaM network and there will be no subsequent flow of data from there. No data will be received or stored by KCL and KCL will not have access to the data. Data will not be transferred to other locations and will not be made available to others within SLaM who are not working on the project described in this application. SLaM will not link the data disseminated by NHS Digital to any other data they may already hold. The HES data being requested will be hosted by SLaM within the Maudsley BRC. Data will be downloaded from NHS Digital with an anonymous, study specific ID and no identifiable information. It will be stored on the SLaM server in an SQL server database on a storage area network and secured using an active directory user group, with access restricted to named individuals according to SLaM’s security policy. Remote access to the database is permitted, but only via a virtual private network accessed using a secure token, such that actual data processing is still carried out on site. Local downloading and printing of data is disabled when it is being accessed via VPN. Who can access the data: Technical assistance with data downloading and processing will be provided by staff within the Maudsley BRC with expertise in database management and information governance who are authorised to access the data for the purpose(s) described, all of whom are substantive employees of SLaM. All other processing activities will be undertaken by the applicant only. The applicant is employed by King’s College London (KCL) and registered as a PhD student within the Department of Psychological Medicine at KCL. They also do clinical work as a psychiatrist for SLaM and has a clinical addendum to their contract with KCL which establishes an honorary contract for this work. It includes provisions to allow sharing of information regarding their performance, conduct and health between SLaM and KCL, a responsibility on the applicant to comply with all relevant NHS policies and procedures and provision for their employment with KCL to be reviewed or terminated if they are in breach of this. Only staff that have received information governance training will be permitted access to the data. What will be done with the data: Analyses will be performed, and results reported at lower super output area (LSOA) level. This is a standard census geography available within the HES data and defined by NHS Digital as a non-identifiable field. Individual level data will be used to allow the standardisation of the LSOA rates calculated for age and sex, both of which are strongly associated with rates of self-harm. Data for analyses will be extracted from the SQL database aggregated into age, sex and year strata within individual LSOAs. The data extracted will be filtered by age and individual diagnostic codes to define cohorts of interest. Similar control/comparison cohorts may also be established. Extracted data will be imported into R, a programme for statistical analysis, from which OpenBUGS, a programme required for the planned Bayesian analyses, will be called. Analyses will be conducted at LSOA level and will involve the calculation of spatially smoothed rates of admission with and without standardisation for age and sex, and adjustment for area level exposures of interest as described in the purpose section above. No record level data will be linked to this dataset, but it will be combined with publicly available data at LSOA level including demographic data to provide denominators for age and sex standardisation, and data relating to the exposures of interest, for example area-level deprivation, area level exposures calculated from census data and area level environmental measures such as air pollution measurements. All outputs are aggregated with small number suppression in line with the HES Analysis Guide. The output of analyses will be descriptive tables for the cohort as a whole, estimates of standardised admission ratios for areas, estimates of the probability that area rates differ from the overall rate for the study area and estimates of the effect sizes for the association between the exposures of interest and self-harm (rate ratios). Thus all will be based on aggregate data. Descriptive statistics will be reported in line with the HES analysis guide, suppressing results if necessary due to cells having low counts. Area-level outputs will be imported into ArcMap, a geographical information system, in order to visualise the results. This will be done at area level with no individuals being identifiable from any of the reported output. Planned analyses: 1) Indirectly 5-year-age and sex standardised rates of all-cause admission and admission for self-harm (ICD-10 codes X60-X84) will be calculated for each LSOA using Bayesian spatio-temporal disease mapping models. The distribution of rates across the Greater London area and over time will be described and compared using overall model parameters, tests for space-time interactions and by mapping rates for different time periods. 2) Sensitivity analyses will be performed for the effect of using broader definitions of likely self-harm. Standardised rates will be calculated for the standard definition of self-harm (ICD-10 Cause codes X60-X84) and broadened definitions including ‘Event of undetermined intent’ (ICD-10 cause codes Y10-Y34), and accident codes that relate to similar mechanisms of injury to those seen in self-harm (e.g. X40-X49, Accidental poisoning by and exposure to noxious substances). Standardised rates , associations with individual and area-level measures and spatial distribution of rates will be calculated and compared using the methods described in 1). 3) Areas with above and below average rates of self-harm admission, before and after adjustment for deprivation will be identified. 4) Predicted rates for the LSOAs will be modelled based on findings from previous work, using publicly available area and population level exposure data. Predicted rates will be mapped and areas predicted to have above or below average rates of self-harm admission identified. 5) Performance of the model in predicting self-harm admission rates will be checked against the clinical data. Reasons for any discrepancies between the model and the clinical data will be explored. What will not be done with the data: * The data will not be linked with 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 except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide. Data required: Only HES APC data is being requested because other datasets do not contain sufficiently accurate and complete coding of self-harm to be used for the study aims. Data is being requested for a 10 year period from 2008-2017. This will allow trends in rates of self-harm by area over time to be examined. Data is only being requested for individuals who were resident within the Greater London region at the time of hospital admission, as this is the study area of interest. Data is only being requested for individuals aged 11 or older at the date of admission in order to match the age range in the data that will be used to build the model being tested. Data for younger individuals is not being used as incidents of self-harm are rare prior to adolescence. Filters have not been applied to specific conditions in order to allow sensitivity analyses of different ways self-harm may have been coded to be carried out, for example how the inclusion of acts coded as of undetermined intent, or accidents affects the results seen and to allow comparison to be made to rates of all-cause admission. No identifiable or sensitive data items have been 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 7 — DARS-NIC-15530-P7L1F

Opt outs honoured: N

Sensitive: Sensitive

When: 2017/06 — 2017/08.

Repeats: Ongoing

Legal basis: Health and Social Care Act 2012

Categories: Anonymised - ICO code compliant

Datasets:

  • Improving Access to Psychological Therapies Data Set

Objectives:

The DoH (2011) strategy document “No Health Without Mental Health” stated that “A priority action for securing improved outcomes is to achieve routine local monitoring of access to services, experience and outcome by sexual orientation”. Assessing referral, access to assessment / treatment, experience and outcomes will help to determine whether service provision is equitable and appropriate for LGB patients. Identifying specific areas of need will highlight where changes are required. This project, conducted by King’s College London (KCL), aims to evaluate treatment access and experience for lesbian, gay and bisexual (LGB) individuals with common mental health problems in relation to Improving Access to Psychological Interventions (IAPT) services in England. The objective is to establish whether there is equitable access and experience for this minority population, and if not, to identify areas are important targets for improvement.

Expected Benefits:

Sexual minority individuals suffer excess rates of mental health problems such as anxiety and depression and are more likely to self-harm or attempt suicide (e.g. Chakraborty et al., 2011; Elliott et al., 2014; King et al., 2008). The NHS Constitution for England states that the NHS "has a wider social duty to promote equality through the services it provides and to pay particular attention to groups or sections of society where improvements in health… are not keeping pace with the rest of the population". The DoH (2011) report “No Health Without Mental Health” identified monitoring of access to services, experience and outcome by sexual orientation as a priority. This project will undertake the evaluation of whether there is equitable access and treatment experiences for lesbian, gay and bisexual individuals in primary care psychological therapies services. This is the first such evaluation of IAPT service provision across England. If inequalities of access or treatment outcomes are identified, the authors will make recommendations about the next steps needed to help improve these health services. The nature of the recommendations will depend on the findings – for example it is possible that particular subgroups, such as sexual minorities who are older or who are also from a minority ethnic background, have reduced access or poorer treatment outcomes. It may be that particular outreach activities to target these groups or additional staff training may be required. The recommendations will be fed back to the National IAPT team who oversee IAPT services in England and the authors will liaise with them about how recommendations may be implemented. They will also be able to assist with dissemination to the local IAPT services. As mentioned above, the findings and recommendations will be disseminated at national conferences attended by IAPT therapists.

Outputs:

KCL plan to publish the results of this analysis in a peer-reviewed journal and disseminate to the relevant health service providers, including the HSCIC. The health services for which it will be most applicable will be Increasing Access to Psychological Therapies services in England. The paper will be published on KCL’s website, subject to any copyright / publishing restrictions by the journal. KCL plan to present the findings at a conference relevant to the health professionals working in these healthcare services, e.g. the annual conference for the British Association of Behavioural and Cognitive Psychotherapies. It is not possible to specify an exact target date as this depends on the time taken to prepare the dataset, analyse the complex data and write up for publication. In all cases, all outputs will be at aggregated level and small numbers will be suppressed in line with NHS Digital guidelines.

Processing:

The data will be used to investigate whether, compared to heterosexual individuals, do LGB people show similar referral levels / pathways, presenting symptoms, treatment access, patient experience and treatment outcomes. KCL aim to investigate whether there is variation depending on the intersection between sexual orientation and other characteristics (e.g. gender, ethnicity or disability, employment status, religion? A further question concerns the disclosure of sexual orientation. This is only recently being collected in IAPT services and KCL will investigate whether willingness to disclose sexual orientation varies according to factors such as age, ethnicity, gender, religion and so on.


Project 8 — DARS-NIC-147847-P6MMR

Opt outs honoured: Yes - patient objections upheld (Does not include the flow of confidential data)

Sensitive: Sensitive

When: 2020/01 — 2020/01.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Identifiable

Datasets:

  • MRIS - Members and Postings Report

Objectives:

The data supplied by NHS Digital to King's Centre for Military Health Research (KCMHR) will be used only for the approved Medical Research Project MR795 - 'Cancer Risk & Mortality in a Sample of Service Personnel Deployed to Bosnia 1992 - 1996'. KCMHR is a research centre within King's College London (KCL). Background: There has been much speculation that military personnel who served on UN peacekeeping duties in Bosnia in 1992-1996 have higher than expected rates of cancer. This speculation has been based on individual case reports and some small clusters of leukaemia reported in Italian armed forces. It has been suggested that the allegedly higher rates of cancer are due to exposure to depleted uranium during deployment (Mayor, 2001). Other European nations have carried out studies to ascertain the incidence of cancer in their Bosnia veterans (Bogers et al, 2013; Gustavsson et al, 2004; Peragallo et al, 2010; Storm et al, 2006). In order to ascertain the rates of cancer and deaths in UK veterans who served in Bosnia, it is necessary to carry out systematic epidemiological studies and to date these have not been done. The study will determine whether the rates of cancer and death among personnel who served in Bosnia are higher than expected. In 1997/8, using information provided by the Ministry of Defence (MoD), King’s College London (KCL) contacted members of UK Armed Forces who had deployed to Bosnia on UN peacekeeping duties as part of a larger study looking at the health consequences of the 1991 Gulf War. Data was collected using a questionnaire. In 1997/8, KCMHR conducted a large epidemiological study on the health consequences of the 1990- 1991 Gulf War (Unwin et al, 1999). This involved identifying a cohort of 4250 individuals who had served in the Gulf, and two comparison groups: 4250 who had served on peace-keeping duties in Bosnia between 1992 and 1996, and 4250 individuals who were in the UK military at the time of the 1990-1991 Gulf War. In addition to demographic, health and service related data, environmental and combat related exposure data were collected from study responders. In 2006, the Gulf War study team supplied personal data (name, date of birth, sex, NHS number, address) of participants in the Bosnia cohort of the 1997 KCL Gulf War study to the Office for National Statistics (ONS) so that participants could be traced and flagged. Self-reported data (and MoD provided data) relating to the Bosnia cohort have been retained as part of the KCL Gulf War study, with data relevant to addressing the aims of this additional sub-study held separately (the Bosnia study database). Until 2014, the Bosnia study database held person identifying data along with cancer notifications and death registration details. On 27/03/2014, all directly identifying data had been removed from the Bosnia database, all hard copy notifications shredded and person identifying details deleted from notifications received electronically. Evidence of this was provided to NHS Digital. The Bosnia study data are pseudonymised. However, the personal data that relate the original KCL Gulf War study are still held at KCL (and will continue to be in anticipation of a possible follow up study). However, these data cannot be accessed by researchers from the Bosnia study. In 2004 ethics committee approval was granted to ‘flag’ the Bosnia cohort with NHS Digital and to obtain cancer registration and death notifications (Joint SLAM/IOP NHS REC, study number 055/04). Section 60 of the Health & Social Care Act 2001 support was applied for and was granted in 2006 to supply patient identifiable data to NHS Digital, to flag the cohort and, for study responders, to supply death registration data, cancer registration data and exit information. For study non-responders summary tables for mortality and cancer incidence would be supplied at the end of the study. NHS Digital was provided with details of the Bosnia cohort (i.e. name, address, date of birth, and NHS number where available). A database of the study responders who were successfully flagged was compiled and is updated with details of cancer registrations and deaths as and when notifications are received from NHS Digital. In 2014 KCMHR again sought and received a favourable ethical review to continue to receive data from NHS Digital (REC reference 14/LO/1141, IRAS project ID 151260). The immediate aim is to ascertain if the number of cancers and deaths reported so far would give sufficient power for meaningful analysis. Once researchers have the latest numbers of cancers and the number of deaths, a power calculation can be carried out to see if there is sufficient power (at least 80%) to show a clinically significant difference between the Bosnia group and a control group. Calculation showed that there were sufficient numbers to show a relative risk of 1.7 in the Bosnia cohort compared to the era group, in other words if there was an increased risk of cancer in the Bosnia cohort of 70% there would be sufficient power to detect it. That was based on the absolute risk of cancer reported by MacFarlane et al who found a low risk (0.5%) of cancer in the era group after 10 years. As time goes on, the absolute risk of cancer in both the era and Bosnia groups will increase (rates of cancers increase with age) so that after ~20 years it would be possible to detect a smaller increase in relative risk. There is not a specific number of cancers that would need to be reached in the Bosnia group since there may not actually be a difference in risk between the two groups. Of more importance is the length of time that needs to have elapsed since exposure for a possible difference in rate of cancer to have occurred. The latency period is the amount of time that elapses between initial exposure and the diagnosis of cancer. The latency period for blood related cancers e.g. non-Hodgkin lymphoma or myeloma have a shorter latency than solid tumours such as lung cancer. For example, the approximate latency period of lung cancer is approximately 14 years, of stomach cancer is 22 years and of kidney cancer is more than 40 years. It is now between 23 and 27 years since the Bosnia cohort deployed and this would be an appropriate time to assess whether the numbers of cancer notifications and /or deaths in this cohort compared to the era cohort show that there is an increased risk attributable to deployment to Bosnia. If there have been sufficient occurrences KCMHR will not require further updates for the cohort. For the longer term and main study aim KCMHR will ask the MoD to supply an anonymised dataset from the flagged “era group” containing information on age, sex, rank and cancer registrations and deaths. Researchers will then be able to compare cancer registrations and deaths in the Bosnia group with a non-deployed group serving at the same time, controlling for age, sex and rank. By comparing to data from the original study (Unwin et al, 1999), researchers will be able to assess whether risk of cancer in the Bosnia cohort is related to reported exposures to potentially harmful materials during deployment. The overall aim of the study is to compare the incidence of cancer in a cohort of UK armed forces personnel who deployed to Bosnia between 1992 and 1996 and a cohort of personnel who were in service at the time but did not deploy to Bosnia. Researchers will also compare the rate of cancer in the Bosnia group with that in the general UK population using publicly available national statistics of cancer and mortality. Additionally, it will be assessed whether the risk of cancer in the Bosnia group is associated with exposures to harmful materials during their deployment. The specific research questions are: 1. Are individuals who served in Bosnia between 1992 and 1996 at greater risk of developing cancer than other military personnel who did not serve there? 2. Are individuals who served in Bosnia between 1992 and 1996 at greater risk of developing cancer than the general UK population? 3. Is risk of cancer in the Bosnia cohort related to self-reported exposures to potentially hazardous materials during deployment? If it is found that military personnel who served on UN peacekeeping duties in Bosnia in 1992-1996 do not have higher than expected rates of cancer, the Ministry of Defence and the Service charities will have evidence to allay fears among this population that they are at increased risk. If researchers find an association between serving in Bosnia and increased risk of certain cancers, heightened awareness among health service providers of this particular risk could lead to earlier diagnosis and improved quality of care. In addition, if researchers find an association between particular exposures and cancer, the Ministry Of Defence could act to protect personnel from such exposures in the future.

Expected Benefits:

Once KCMHR have sufficient data for analysis there will be potential benefits to the population of UK Armed Forces personnel who deployed on UN Peacekeeping operations in Bosnia. If it is found that military personnel who served on UN peacekeeping duties in Bosnia in 1992-1996 do not have higher than expected rates of cancer, the Ministry of Defence and the Service charities will have evidence to allay fears among this population that they are at increased risk. There has been speculation of such an increase based on some small clusters of leukaemia reported in Italian Armed Forces personnel. Exposure to depleted uranium has been suggested as a possible risk factor. Conversely, if researchers find an association between serving in Bosnia and increased risk of certain cancers, heightened awareness among health service providers of this particular risk could lead to earlier diagnosis and improved quality of care. In addition, if researchers find an association between particular exposures and cancer, the Ministry Of Defence could act to protect personnel from such exposures in the future.

Outputs:

The key milestone is the decision in if there is sufficient data to carry out significant analysis immediately or whether there is a need to wait a few years until further data is available. This is a long term study recording the cancer status of veterans of The UK Armed Forces deployments to Bosnia between 1992 and 1996. In view of the anticipated long latency period between exposure and the appearance of cancers there may not be sufficient data yet to carry out the analysis. Once a decision has been made that there is sufficient data the outputs will include: - a report of the findings from the study with recommendations to the UK Ministry of Defence (MoD) and to Service charities dealing with veterans and their families (for example, The Royal British Legion, SSAFA). The report will include a lay summary which can be made available on the webpages of the MoD and Service charities. - an academic paper giving an analysis and discussion of the results of the study that, subject to acceptance, will be published in an appropriate journal such as the BMJ. - an infographic will be produced which will be hosted on the King’s website and distributed to all stakeholders for onward dissemination. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

Processing:

In 2006 name, address, date of birth, and NHS number of the sample of UK Armed Forces personnel, who had deployed to Bosnia between 1992 and 1996 were sent from the KCMHR to the ONS so that they could be identified and flagged for long-term follow-up. Of the 4500 people sampled into the Bosnia group, 2620 responded to the questionnaire study. 2470 of those who responded were successfully flagged, and any death and cancer notifications or embarkations have been sent by ONS and subsequently by NHS Digital (after the service transferred in 2008) to KCMHR for this cohort. The data received under this Agreement are pseudonymised - i.e. personal identifiers such as name, date of birth and NHS number are not sent to KCMHR. The pseudonymised notifications from NHS Digital are downloaded and entered onto an analysis database by a Research Associate at King’s College, London. The database held by KCMHR is encrypted and password protected using 256-bit AES. It is accessed on a networked PC which is connected to a restricted access server at King’s College London, Strand campus and ‘backed up’ at Jisc Shared Data Centre, Slough (hosted by Virtus Data Centre). Access to the server is restricted to a single research associate. There are no personal identifiers held in the database. The administrative database which contains the identifying details of the cohort with a common unique identification number which could technically re-identify individuals in the analysis database, has been archived to an encrypted external drive and is stored in a Ministry Of Defence approved safe. The combination to the safe is known to three people only: the Project Manager, the Database Administrator and a KCMHR research associate. Access to the safe is logged with the name of the person, date and reason for accessing. The research associate does not have access to the password to the encrypted external drive which is held in a secure password vault. No one who can access the administrative database can access the analysis database. These technical and organisational controls ensure the data under this Agreement remains pseudonymised. The data will be analysed under supervision by KCMHR's Chief Investigator. Cancer and mortality rates will be calculated for the Bosnia group and compared with a group that was in service at the same time but did not deploy to Bosnia. A comparison of the rates of death and cancer will also be made with the UK general population. Exposures to environmental hazards will be examined to look for any associations with cancer. Only the Bosnia group from the Gulf War study were flagged for KCL. However, the Gulf War details of all personnel who served in the Gulf area (n=53,462) were sent to ONS by the MoD along with an equal sized cohort of personnel who were in service but did not deploy to the Gulf - termed the ‘era’ group. Over 96% of both cohorts were identified and flagged on the NHS central register. KCL will apply to Defence Statistics at MoD for a pseudonymised dataset of their era cohort containing information on age, sex, rank and cancer registration or deaths and, subject to access to this data being granted, this will be used as the ‘control’ group. Statistical analysis will be carried out using the statistical package, Stata. KCMHR will carry out a survival analysis, the outcome being defined as incident cancer registrations or death from cancer. For the Bosnia cohort, time at risk will be calculated from the point when the individual returned from deployment. For the control group, time at risk will be calculated from 1st April 1991. Failure events will be cancer registrations or deaths (whichever is recorded first). Individuals who emigrated from the UK will contribute up to the point of migration. Cox's proportional hazards model will be used to calculate hazard ratios with 95% confidence intervals for cancer registration. KCMHR will control for sex, age and rank. Standardised incidence ratios (SIR) and standard mortality ratios (SMR) will be calculated to compare cancers and deaths in the Bosnia group to the UK population. Multivariable logistic regression will be used to examine associations between cancer and environmental exposures adjusting for potential con-founders restricted to the Bosnia cohort. The key milestone is the decision in if there is sufficient data to carry out significant analysis immediately or whether there is a need to wait a few years until further data is available. This is a long term study recording the cancer status of veterans of The UK Armed Forces deployments to Bosnia between 1992 and 1996. In view of the anticipated long latency period between exposure and the appearance of cancers there may not be sufficient data yet to carry out the analysis. Once a decision has been made that there is sufficient data the outputs will include: - a report of the findings from the study with recommendations to the UK Ministry of Defence (MoD) and to Service charities dealing with veterans and their families (for example, The Royal British Legion, SSAFA). The report will include a lay summary which can be made available on the webpages of the MoD and Service charities. - an academic paper giving an analysis and discussion of the results of the study that, subject to acceptance, will be published in an appropriate journal such as the BMJ. - an infographic will be produced which will be hosted on the King’s website and distributed to all stakeholders for onward dissemination.


Project 9 — DARS-NIC-144761-Y3X9Y

Opt outs honoured: No - data flow is not identifiable (Does not include the flow of confidential data)

Sensitive: Sensitive

When: 2019/11 — 2019/11.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant

Datasets:

  • Hospital Episode Statistics Accident and Emergency
  • Civil Registration - Deaths
  • Hospital Episode Statistics Admitted Patient Care
  • HES:Civil Registration (Deaths) bridge

Objectives:

King’s College London (KCL) requires the data necessary for its research project ‘End of Life Care Outcomes for Adults with Serious Mental Illness (SMI)’. The aim of this work is to explore the end of life care circumstances of adults with a diagnosis of serious mental illness (SMI). End of life care outcomes are measured by hospital admissions and A&E visits at the end of life and place of death. The project aims to achieve a clear picture of where patients with SMI die and their health care utilisation at end of life and assess what demographic and/or clinical factors are associated with place of death in this patient group. The objectives are as follows: • To describe the acute care service use (including hospital admission, the length of hospital stay, A&E visit) in the last year of life, and place of death in people with SMI • To evaluate factors associated with acute care service use and place of death in those with SMI and their relative importance • To explore the relationship of the acute care service use and place of death. This project is in the public interest as KCL believes it is making a valuable contribution to science and, with the results, KCL hopes to demonstrate whether or not patients with SMI face inequalities in care at end of life. Ethical approval has been obtained. There is minimal risk of harm as data will be pseudonymised, stored securely and summary data only will be reported in dissemination. The data requested are required to meet the following three aims: 1] To describe the acute care service use (including hospital admission, the length of hospital stay, A&E visits) in the last year of life, and place of death in people with SMI. To meet this aim KCL requires the following items from HES data: the dates of admission and discharge to generate measures of number of admissions/A&E visits and the length of hospital stays in the last year of life, and mortality data, which includes individuals’ place of death. Place of death can be summarised as hospital, home, hospice, care home and other, if the place of death could potentially identify an individual (e.g. a residential address). These data will enable KCL to describe the outcomes of interest for this project. 2] To evaluate factors associated with acute care service use and place of death in those with SMI and their relative importance. This aim will identify the demographic and clinical factors associated with the outcomes of interest: acute care at end of life and place of death. Patient demographics and clinical factors, available from MHDS and mortality data, are required to explore associations with the outcomes derived from HES and mortality data. 3] To explore the relationship between acute care service use and place of death. To meet this aim, KCL will use the derived measures of number of admissions and length of hospital stay and number of A&E visits (generated from HES data) and the categorical measure of place of death (from mortality data) to explore bivariate associations between these outcomes. One member of the research team applied for and secured funding from KCL to undertake this project. This was to continue and expand upon previous working carried out at a local level. Previous work by the research team has explored end of life care outcomes in adults with SMI in a south London dataset (Wilson et al, 2019*). This work identified inequalities in end of life care in this localised patient group; the purpose of the proposed project is to assess end of life care at a national level. The research team have also conducted a systematic review (unpublished, currently under review) assessing the current evidence base for the end of life care outcomes for people with SMI. This review found that more focused, rigorous research is needed in this area. *Wilson et al. 2019. “Place of Death and Other Factors Associated with Unnatural Mortality in Patients with Serious Mental Disorders: Population-Based Retrospective Cohort Study.” BJPsych Open 5(2): e23. This project builds on an ongoing project assessing end of life care in patients with SMI in South London, which uses a linked dataset from the South London and Maudsley Trust. For clarity, this project will not involve linkage with or any other use of data from that other project. Under this Agreement, KCL requires data from a one-year time period. If this project proves an efficient way of demonstrating end of life care outcomes in patients with SMI and highlights inequalities in this population, the research team may request data from a long period of time to perform longitudinal analysis and explore time trends. The results of this project may contribute to/inform the development of a post-doctoral fellowship application. This is a cross-sectional observational study including a cohort of people with SMI. The sample are adults (18 years+) who died with a diagnosis of schizophrenia, schizotypal or delusional disorder (ICD 10 F20-F29), bipolar disorder (F31), depressive episode (F32) or recurrent depressive disorder (F33). There will not be a control group. KCL requires Hospital Episode Statistics (HES) and mortality data for use in the study, End of Life Care Outcomes for Adults with Serious Mental Illness. Linked data for eligible observations are required at an individual level for individuals that died within one calendar year (the latest year for which all linked data are available). Data will be pseudonymised. Mortality data are required to assess place of death and cause(s) of death. HES data are required to identify patients with a diagnosis of SMI (defined above) who died within the HES 2018/19 year. HES data (admissions and A&E) are also required to measure acute service use in the last year of life. Data are required for one year to measure admissions and A&E visits in the last year of life, the time frame considered to capture health care use at the end of life. National data are required as this proposed project is a continuation of a previous project that explored end of life care outcomes in patients with SMI in south London (data were provided by a south London mental health trust, SLAM). National data are required to expand on this localised project. Data are requested for patients with the following psychiatric diagnoses: schizophrenia, schizotypal or delusional disorder, bipolar disorder, depressive episode or recurrent depressive disorder. Date of birth fields are to be provided in Month/Year format for pseudonymisation. Data are requested only for eligible individuals (those with an eligible diagnosis). KCL are requesting data for just individuals who died within HES year 2018/19. KCL is the only organisation involved in this study and the sole organisation responsible for determining how and why the data will be processed. All researchers involved in the study are employed by KCL as research or academic staff.

Expected Benefits:

Dissemination in peer reviewed journals and at conferences and the dissemination of the results through the KCL network (which includes collaboration with the South London and Maudsley NHS Trust, Europe’s largest provider of mental health care) is intended to change policy and directly influence adult health care for adults with SMI. Dissemination of results is in the public interest and the study team intend to use the data to show whether or not there are inequalities in care for people with SMI. The study team will be submitting our results to open-access high impact journals. Open-access is crucial for dissemination of the study results as KCL hopes to reach as wide an audience as possible. Publishing in high impact journals is also important to represent the value of the work. The main target journal will be the British Journal of Psychiatry, as a high impact journal within the field of psychiatry (impact factor 5.9; 10 out of 142 for psychiatry) and the primary journal for psychiatrists in the UK. Failing publication in the British Journal of Psychiatry, alternative journals to consider would be The Lancet Psychiatry and the Journal of Psychiatric Research. The results will be presented to psychiatrists and members of mental health teams at the Institute of Psychiatry at King’s and share our results with other clinical teams. the study team will pursue opportunities to present the results at clinical forums, such as the European Psychiatry Association congress and the European Association of Palliative Care annual conference. The study team will produce policy briefs summarising the results of the study. These will be used to respond to any relevant calls for information put out by MPs, health care committees, task forces or special interest groups. The study team will proactively look for any individuals or groups (in policy) with an interest in mental health and health care. The study team will also disseminate our policy briefs at study days and events held in our department, which are regularly attended by policy makers. The outputs (peer reviewed publications, presentations at scientific forums, policy brief) achieve the purpose of the project by disseminating the results we find from the data. The benefits of the dissemination plan include targeting and reaching the most relevant individuals. Papers will be sent to psychiatry journals and abstracts to psychiatry and social science conferences. Policy briefs and results summaries will be shared with relevant charities and third sector groups. The outputs from this project will show, for the first time as far as the study team are aware, what acute health care adults with SMI in the UK receive in the last year of life and where they die, demonstrating whether there are inequalities in place of death compared to the general population. KCL will also be able to show if other factors (e.g., age, diagnosis, gender, ethnicity) are associated with inequalities in health care access. This evidence is needed for change (in the provision of care for this patient group). The results. Through publishing in a psychiatry journal, health care professionals are more likely to see these results and be made more aware of the end of life care issues faced by their patients, this could help to change practice. If the results of the project are picked up by policy makers, they could potentially inform policy around the provision of end of life care for patients with SMI. The benefits will be measured in several ways. First, the reach of the results will be measured by summarising and evaluating the dissemination of the work, including readership of the journal in which the results are published, number of downloads, online readers and retweets. The approximate size of audiences to which the results are presented will be measured. Impact of the results will be measured initially by engagement with clinicians and policy makers. Raising awareness of end of life care in patients with SMI in clinicians (palliative care, psychiatry and general medicine), though not measurable, is an essential benefit for initiating change in the provision of end of life care for patients. The goal of this work is to raise awareness of patients’ end of life care needs in clinicians working in psychiatry and to raise awareness of an underserved population (patients with SMI) in palliative care for relevant health care professionals. Discussion between partners and collaboration between palliative care and psychiatry will be an indicator of education and awareness of the project and research goals. It will be possible to measure direct benefit for patients when changes are made to reflect the increased awareness of the need for high quality end of life care, although this is likely to take time. These changes will include more focus of advanced planning for patients with SMI and recording end of life care preferences, less emergency care at end of life (which will be a marker of better end of life care planning) and more patients accessing high quality palliative care at the end of life. These outcomes will be measurable in future years when data are available to depict what end of life care patients are receiving. In future years, KCL will be able to perform time-trend analysis to explore if outcomes have improved over time. KCL hopes that its research ultimately improves end of life care for people with SMI in 2014 0-3.3% of the population in England are estimated to have a diagnosis of SMI (roughly 1.0m people). This is an under-researched area of health care provision. There are concerns that people with SMI have received worse health care at end of life and this work will demonstrate whether these patients experience health care inequality at the end of life. If the data shows that people with SMI have worse health care experiences, this provides evidence of inequality and support for improvement of services (including better end of life care) for patients with SMI. Change in provision of services or prioritisation of a vulnerable patient population is not possible without robust evidence to demonstrate need.

Outputs:

The following outputs will be produced: Progress reports will be provided to the funders (King’s College London) throughout the project. These will describe the progress of the work (e.g. having received the data, analysis, manuscript writing) and will not divulge any data/results. The first output will be a short report, published in a psychiatric journal, by the end of the year (December 2019). This output will report descriptive statistics for the cohort, including place of death and cause of death and other demographics. The main results of the paper will be published in a psychiatric journal in 2020. This paper will report the factors (demographic and clinical) associated with place of death in patients with SMI. It is hoped that this will demonstrate the pathways by which various demographic variables are associated with health care and end of life care. By publishing in a psychiatric journal, psychiatrists and mental health professionals are the target audience, thus the aim of this work and targeting this audience is to highlight the importance and needs for good end of life care in this vulnerable patient group. Additional outputs will include abstracts of the main results to one or more of the following: the European Psychiatric Association in 2020, the European Association for Palliative Care in 2020 or the Society for Social Medicine 2020. Results of the study will also be disseminated to patient and carer groups, including Mind and MQ Mental Health and to followers of the work of the Cicely Saunders Institute, through Twitter, PPI groups, etc. KCL will also write a policy brief, which will summarise the work and the main results in a policy brief and will be disseminated to relevant groups and individuals. Outputs will always be aggregate with small numbers suppressed. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

Processing:

NHS Digital will create the cohort using HES filters and extract relevant records and pseudonymise the data. There will be a single flow of linked and pseudonymised record level data from NHS Digital to King’s College London (KCL), with pseudonymised ID and no identifying data. There will be no further flow of data. KCL will be the sole organisation involved in processing the data. As per the application purpose section, researchers at KCL will manage the storage, cleaning, analysis and interpretation of the data. The data will not be linked with any record level data or be matched with publicly available data. There will be no requirement or attempt to re-identify individuals from the data. Data will only be accessed by individuals within the Cicely Saunders Institute, KCL who have authorisation from NHS Digital to access the data for the purpose described, all of whom are substantive employees of KCL. KCL will store the data on a secure server at the Cicely Saunders Institute, KCL, which can only be accessed within the department (located at Denmark Hill campus, KCL) by researchers named on the project. Data analysis will be conducted by role based access, limited to researchers working in the study team, on the departmental computer on which the data are stored (ie, analysis will not be conducted remotely). Data will be accessed by employees on a secure server on a desktop computer within the department (Cicely Saunders Institute, KCL). The data will not be made available to any third parties. Results will be presented at an aggregate level in research outputs. All data will remain anonymous. Small cell counts will be suppressed (N<10).


Project 10 — DARS-NIC-134027-L9T9J

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

Sensitive: Non Sensitive, and Sensitive

When: 2019/12 — 2019/12.

Repeats: One-Off

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

Categories: Identifiable

Datasets:

  • Hospital Episode Statistics Accident and Emergency
  • Civil Registration - Deaths
  • HES:Civil Registration (Deaths) bridge

Objectives:

The purpose of the current project is to investigate the longer-term outcomes of treatment engagement and A&E attendance among adolescents accessing South London and Maudsley Children and Adolescent Mental Health Services (CAMHS) who received a 'Therapeutic Assessment' (TA) versus 'Assessment as Usual' (AAU) when they initially presented to A&E for self-harm. This is an 8 year follow up study from an original randomised controlled trial which demonstrated that TA, a brief, psychotherapeutic intervention for adolescents who present with serious self-harm, significantly improves engagement with follow-up treatment (Ougrin et al., 2011). This model has been independently replicated by a US research group with remarkably similar results (Asarnow et al., 2011). The first follow-up study showed that the impact of TA on engagement persists at least 2 years after the initial intervention (Ougrin et al 2013). This had led to a recent update with a recommendation for TA in the NICE guidelines for self-harm in adolescents (https://arms.evidence.nhs.uk/resources/hub/964618/attachment). There is significant interest in training and implementation of TA both in the UK and around the world. The chief investigator’s research group developed the model of TA and undertook its evaluation. An aim of the current study is to further investigate whether the treatment engagement benefits observed at 2 years follow up still exist at 8 years. In addition, the research wishes to identify whether the use of TA had any positive effect on A&E presentations for self-harm in the longer-term. Outputs provided from NHS Digital will therefore feed into a larger study which also explores the level of treatment engagement among individuals who have continued to access South London and Maudsley NHS services. Both South London and Maudsley NHS Foundation Trust (SLaM) and King’s College London are hosting this research project in order to collect the relevant patient information and analyse the aggregate data, respectively. No elements of the work will be taking place outside the UK. The same applies for existing data collected by the team. Aggregate-level data will be published in a peer reviewed journal.

Expected Benefits:

The data provided will contribute to research which aims to identify whether there are any long-term benefits to using a therapeutic assessment model for self-harm in adolescents accessing mental health services. This will yield additional insights to the findings already observed at short and medium-term time points. To date, the existing research has led to a recent update with a recommendation for TA in the NICE guidelines for self-harm in adolescents (https://arms.evidence.nhs.uk/resources/hub/964618/attachment). Given that this study is the first of its kind to investigate long-term outcomes for self-harm focused intervention, the research team hope that the research will be of benefit in the following ways: - To help inform psychosocial assessment protocols for this group of patients - To contribute to a 3-year programme of work which seeks to improve the quality of patient care within the trust - To update and inform further training in Therapeutic Assessment for mental health professionals both within the UK and worldwide.

Outputs:

Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide. Aggregate level data will shared in the following ways: - Publishing in an open access peer reviewed journal. The original study has been published in Archives of Disease in Childhood. This will therefore be accessible to academics, clinicians and members of the general public - Self-harm research meetings within South London and Maudsley NHS Foundation Trust (open to Children & Adolescent Mental Health Service academic and clinical staff) - Quality Improvement meetings within South London and Maudsley NHS Foundation Trust. Led by senior clinicians as part of a three-year programme aiming to introduce improvements that will drive up the quality of patient care Deadlines: data on treatment engagement was collected in September 2017. The research team aim to collate all data by the end of 2018.

Processing:

SLaM will submit identifiers of the cohort to NHS Digital for linkage to pseudonymised HES and Civil Registration (mortality) data. Data on the cohort's A&E attendance rates will be obtained from NHS Digital containing the Study ID which will be received by a member of the research team based in the Supported Discharge Service (Maudsley Hospital) where it will be stored securely on South London and Maudsley NHS computers. The data received will be linked with the existing patient data on basic demographics and treatment engagement. For this reason s251 approval has been sought and granted by CAG. Any identifiers will then be removed before the data is transferred to members of the research team and statisticians within King’s College London for analysis. Data will be aggregated for publishing. No identifiable information shall be stored outside of NHS trust computers. Collection of data via this means has been approved by the Research Ethics Committee with section 251 support (REC reference: 16/EE/0308). Data will be kept securely in accordance with South London and Maudsley NHS Foundation Trust’s and King’s College London’s data protection policies, which research staff have knowledge of from mandatory training. All researchers working on the study are directly employed within King’s College London or South London and Maudsley NHS Foundation Trust. Taking the short to medium-term findings observed from the original study and two year follow-up, it would be important from the scientific and clinical points of view to find out if the dichotomy between treatment engagement and A&E attendance rates remains in the longer-term or if the improved engagement leads to changes in A&E attendances. 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). All outputs will only contain results in highly aggregated format and as statistical summaries and measures of association. Small numbers will be suppressed in line with the HES Analysis Guide. Record level information will not be released to any third party.