Request for ONS linked data for a cohort of patients diagnosed with Beta-thallasaemia and myelodysplasia syndrome between 01/01/2015 to 31/12/2019.

Health IQ Limited (Health IQ) is a research consultancy specialising in providing research and analytics expertise to healthcare and life sciences. Their consultants have experience in the application of data to solve specific problems within the industry. Their work spans primary, secondary, mental health, community and wider societal datasets. Health IQ Limited produce tools and reports used by health and social care to achieve the following broad aims:

1. Understand and quantify the burden of disease.

2. Support service improvement in terms of treatment and efficiency of service.

3. Add to the body of healthcare knowledge available through robust research”

For this particular project, Bristol-Myers Squibb (an American multinational biopharmaceutical company and a client of Health IQ) have funded Health IQ to undertake a one-time analysis and provide a report looking at the two conditions below:

Beta-thalassaemias are a group of hereditary blood disorders characterized by anomalies in the synthesis of part of red blood cells that carries oxygen throughout the body resulting in issues ranging from severe anaemia to other medical conditions while myelodysplastic syndromes (MDS) are an often unrecognized, under-diagnosed rare group of bone marrow failure disorders, where the body no longer makes enough healthy, normal blood cells in the bone marrow where a risk of progression to acute myeloid leukaemia (AML) exists.

Although, the prognosis of individuals with beta-thalassaemia has substantially improved in the last couple of decades as a result of advances in treatment methods, there is a gap in the knowledge of what a typical patient journey looks like in the actual hospital setting.

This study is therefore aimed at mapping out the patients journey of this cohort of patients from diagnosis including the treatments received, clinical outcomes e.g. death and the associated resource usage using secondary care data collected in England.

This is a retrospective observational study on an administrative healthcare data set in England between 01/01/2015 and 31/12/2019 with no link to other studies. The cohort was identified using ICD-10 codes from Hospital Episode Statistics (HES) data during the study period . The original HES data used to identify the cohort was obtained under agreement DARS-NIC-15293-R6V2H between Health IQ and NHS Digital. Under the original agreement Health IQ have access to HES Inpatient, Outpatient and Accident and Emergency (A&E) datasets which enabled them identify patients with the two conditions as those who had ever been coded within the preceding 5 years. The endpoints to be measured cover demographics and clinical patient profiles (including age, gender, comorbidities), treatment patterns (treatment specialities) and healthcare resource use (inpatient admissions, length of hospital stay, outpatient appointments, A&E attendances and costs). As an observational study, there will be limited utility establishing causality thus Health IQ will not be able to establish the statistical relationship between the conditions and selected endpoints, although this is not an objective of the study. The objectives are thus primarily descriptive as below.

Objective 1 - To describe the patients diagnosed with beta thalassaemia or MSD in terms of demographics and clinical characteristics.

Objective 2 - To describe the patient’s treatment pathway from diagnosis including procedures, transfusions and the specialities they have been treated in.

Objective 3 - To quantify the associated healthcare resource use and tariffs for this patient cohort including inpatient, outpatient, and A&E care.

Objective 4 - To quantify the clinical outcomes associated with patients with beta thalassaemia including adverse events resulting from blood transfusions, and death.

Objective 5 - To describe mortality outcomes such as cause of death and time to death (survival analysis).

For objective 5, although HES data has death records (mortality data), it is only recorded when a patient dies during a hospitalisation which is not sufficient for to achieve the study objectives. As mortality is one of the end points to be determined to achieve one of the objectives of this study, Health IQ therefore require Civil Registration mortality to ascertain when patients within the cohort die and what the recorded cause of death is outside of hospital. This information is only available in the Civil Registration (Deaths) data extract and hence why Health IQ require linkage.

Under the original agreement DARS-NIC-15293-R6V2H Health IQ's GDPR Legal Basis for processing the pseudonymised record-level data from NHS Digital was Articles 6(1)(f) Legitimate Interests and 9(2)(j). Health IQ wish to continue to process the data in order to conduct analysis that will support the optimal delivery of healthcare as a whole, and hence be a benefit to the individuals and all patients. Patient-level data is required in order to accurately conduct the above types of analysis, without this Health IQ could not conduct the level of research needed to provide new insights into diseases or treatment pathways, and the potential benefit to healthcare of this research would not be realised. Health IQ have carried out a Legitimate Impact Assessment using the Information Commissioner’s Office (ICO) template and has been reviewed by NHS Digital. NHS Digital continue to conclude that Health IQ can rely on legitimate interests for this processing.

However, this agreement requests only pseudonymised record-level data on Mortality. Date of death is being released under this agreement and NHS Digital have satisfied themselves that this will not render the dissemination identifiable, and that the dissemination remains pseudonymised and is not owed a duty of confidence.

Only aggregated data with small number suppression applied as per the HES analysis guide will be disseminated as part of a study report.

Health IQ will be the sole Data Controller and Funder for this request, who also processes the NHS Digital data. In this agreement Amazon Web Services (ASW) is, strictly, a Data Processor in the sense that NHS Digital data are hosted and manipulated on their infrastructure. By design, AWS themselves cannot access or read any of the NHS Digital data in Health IQ that are hosted on their infrastructure, nor can anyone else who is not specifically granted individual access to NHS Digital data (including Health IQ substantive employees).

The cohort will be identified from the NHS Digital pseudonymised HES data by Health IQ based on ICD-10 codes using agreement DARS-NIC-15293-R6V2H. Each individual record is assigned a unique HES-ID.

The cohort of will be shared with NHS Digital to be linked to the Civil Registrations (ONS) data set using the HESID field to identify the required mortality data. Linkage will be based on unique HESIDs which will ensure no duplication can occur.

The extracted record-level pseudonymised data (containing only those records where a death is registered) will then be sent to Health IQ by NHS Digital through a secure file sharing mechanism called SEFT (Secure Electronic File Transfer).

Health IQ will then process the record level pseudonymised data to carry out survival analysis and determine cause of death for each individual record. The results will be aggregated with small number suppression applied for the whole cohort and presented in a study report.

Data will be stored in a secure cloud platform – Amazon Web Services and are only accessible through secure access 2-factor authentication mechanisms (i.e. via a Virtual Private Network (VPN)). Analysis will only be carried out by Health IQ analysts who are all substantive employees who have monitored access to the databases. Health IQ staff will carry out the analysis using remote devices that are secured through very strict access password-protected 2-factor authentication access system. All staff who will access the data have regular IT/ Information Governance/ Data Protection training. No data is permitted to be downloaded and stored on remote devices. All storage and processing will be conducted in the cloud server platform.

Amazon Web Services is, strictly, a data processor in the sense that NHS Digital data are hosted and manipulated on their infrastructure. By design, AWS themselves cannot access or read any of the NHS Digital data in Health IQ that are hosted on their infrastructure, nor can anyone else who is not specifically granted individual access to NHS Digital data (including Health IQ substantive employees). Therefore, any access to the data held under this agreement would be considered a breach of the agreement. This includes granting of access to the database[s] containing the data.

Amazon Web Services UK are compliant with standard security frameworks, including ISO 9001, 27001, 27017, 27018; the Cloud Security Alliance certification and UK Cyber Essentials Plus. Amazon Web Services will only be storing NHS Digital data on UK based servers.

Processing of the data is restricted to within England and Wales. No attempt to re-identify individuals will be undertaken by Health IQ Limited.

DISCLOSURE CONTROL / SMALL NUMBER SUPPRESSION

In order to protect patient confidentiality, when presenting results calculated from record level Mortality data, outputs will contain only aggregate level data with small numbers suppressed in line with HES Analysis Guide. When publishing Mortality data, you must make sure that:

· cell values from 1 to 7 are suppressed at a local level to prevent possible identification of individuals from small counts within the table.

· Zeros (0) do not need to be suppressed.

· All other counts will be rounded to the nearest 5.

Data will not be made available to any third parties other than those specified except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

The output from this data processing is expected to be an excel report and a peer reviewed article that will be submitted to a peer reviewed journal such as the British Medical Journal (BMJ) or Lancet. The reports are hoped to also be used to submit poster and oral presentations to relevant conferences e.g. International Conference on Immunological, Hematological and Rheumatological Sciences. It is expected that this will enable the results of this study to reach policy makers such as the NHS, commissioners, disease area experts, clinicians, other researchers, scientists, clinician groups and pharma. The study team hope to influence policy for care of patients with these conditions, inform budget allocations and foster engagement on issues affecting patient care within this disease area thus ensuring the knowledge generated is of public health benefit for the UK and the rest of the world.

Health IQ Limited aim to have the information generated to be communicated through the different channels between July and December 2021.

The reports will be aggregate data with small numbers suppressed based on HES analysis guidance to make sure deductive disclosure does not happen. No individual level data will be published or disseminated to the public.

NHS Digital will be recognised as a data provider in all reports or articles produced as part of this data processing.

This study is aimed at generating knowledge to understand the treatment pathway, healthcare resource use and costs and clinical outcomes associated with the Beta thalassaemia and myelodysplastic syndromes. Health IQ would also want to look at whether there is a relationship between transfusions and survival and whether the degree or increase in transfusion over time change survival in these patients. Currently there is a lack of data on treatment pathways, healthcare burden and the impact of survival on UK patients with beta thalassaemia and MDS and how this compares to the general population. It is believed that these patients have a worse prognosis and therefore having greater understanding of this may help in shaping the treatment pathway in the future.

The knowledge generated once disseminated with the intended audiences through the suggested channels publication and presentation at conferences is hoped to go a long way within informing policy makers, clinicians and drug manufacturers of the real-world picture in relation to these two conditions. This, in turn, is hoped to inform budgeting, clinical management decisions and support evidence generation for Reblozyl[a] for the European Medicines Agency (EMA) review for the treatment of MDS and Beta thalassaemia patients with ring sideroblasts which would all lead to improved patient experiences and outcomes in the future once approved.

All output will be generated by Health IQ and the evidence will be shared with Bristol-Myers Squibb. The Main purpose of this analysis would be to generate evidence for Reblozyl for EMA review for the treatment of MDS and Beta thalassaemia patients with ring sideroblasts. As a result Bristol-Myers Squibb will use this data to inform their engagement with regulatory and licencing authorities. Bristol-Myers Squibb is the manufacturer of Reblozyl and as such would stand to gain by way of commercial profit if the drug is adopted for use in the UK and in other healthcare systems around the world.

[a] https://news.bms.com/news/corporate-financial/2020/European-Commission-Approves-Reblozyl-luspatercept-for-the-Treatment-of-Transfusion-Dependent-Anemia-in-Adult-Patients-with-Myelodysplastic-Syndromes-or-Beta-Thalassemia/default.aspx

Health IQ aim to have these findings published, targeting high impact journals (e.g. the BMJ, Lancet, NEJM, Value in Health). The resulting publication shall follow relevant guidelines for each journal, as well as the principles outlined in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Understanding how these two conditions are managed for patients in the health system, the usage of resources and what happens to these patients post treatment including survival is hoped to lead to:

i. Improved treatment options and policy that is based on data.

ii. Data led resource allocation and treatment strategies for any patient diagnosed with these conditions.

iii. Data le assessment of unmet need to inform the need for an intervention of the most at risk patients.

iv. The above will all lead to better treatment outcomes hence reducing preventable morbidity and mortality due to these conditions amongst the population.

This is hoped to be of significant benefit to newly diagnosed and existing NHS patients with beta-thalassaemia and myelodysplasia syndrome.

Yielded Benefits is not a requirement for new applications.