NHS Digital Data Release Register - reformatted
CSL-UK (CompuFile System Limited)
Project 1 — DARS-NIC-01207-V9G9P
Opt outs honoured: No - data flow is not identifiable (Does not include the flow of confidential data)
Sensitive: Non Sensitive
When: 2016/04 (or before) — 2019/10.
Legal basis: Health and Social Care Act 2012, Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)
Categories: Anonymised - ICO code compliant
- Hospital Episode Statistics Admitted Patient Care
- Hospital Episode Statistics Critical Care
- Hospital Episode Statistics Outpatients
Some examples of benefits delivered within the past year: 1) Nice Quality standard [QS68] targets the healthcare system to deliver “Coronary angiography and PCI within 72 hours for NSTEMI or unstable angina”. This target is in place to reduce the risk or a further cardiovascular event, and ultimately to improve life expectancy. CSL has provided some extensive analysis to a cardiologist working for a local trust to help measure the effect of a pathway redesign they have implemented recently on the number of patients meeting this target. Using these data, they have repeated the analysis for three other NHS Trusts, enabling the cardiologist to work with these trusts to identify areas of performance that can be improved following the experience of the initial Trust. Approximately 1/3 of the 70,000 or more patients received an angiography or PCI outside of the recommended time last year, so improvements in this area can have a large impact on patient survival and quality of life. 2) CSL has worked with a CCG over a number of years to provide data and insights to aid its program of redesigning its COPD services. This was initiated following the CCG’s identification of this area as a high priority for improvement of patient care. Subsequently the CCG has achieved a reduction in emergency admissions for COPD of over 2%. This came at a time when the national level of admissions were rising at 1%. This means potentially 100 emergency admissions being avoided by the CCG per annum, saving in the region of £200,000 pa. In the latest data they have the proportion of COPD admissions that are an emergency in this CCG are approximately 10% lower than the national average.
Compufile Systems Ltd (CSL) is a data intermediary, and has been helping healthcare organisations understand and process data for over 25 years. As such CSL take data security very seriously, are ISO27001 certified, and regularly penetration test these systems. CSL process non-sensitive, pseudonymised HES data to provide third parties involved in the delivery of healthcare with aggregated, suppressed data (small numbers not shown), and provide expertise and advice to help them interpret these numbers. Customers CSL’s services are offered to a variety of organisation types involved in the provision of healthcare to patients: • NHS organisations (Type 1), made up of Clinical Commissioning Groups, Clinical Support Units and Hospital Trusts only. A basic service is provided free of charge to NHS organisations, though subject to the same purpose limitations and legal agreements set out below. • NHS suppliers (Type 2), including medical device companies and life science companies to carry out the functions included in any contracts/commissioning from NHS organisations, or to support initiatives to deliver cost savings or quality of care improvements to their NHS customers (usage is not permitted for solely commercial purposes). Further detailed explanations of the purposes for which the two types of organisation use the data are provided below, with examples given. Data minimisation CSL request the full breadth of data within HES in order to provide analysis to a number of different organisations with different areas of interest – both with regard to geography and specialty. However, the data provided to NHS suppliers is limited to the specialty areas relevant to the use specified in their contracts with CSL. Access for NHS Customers is limited to the data for the Trusts & CCGs relevant to their needs (typically data for their Trust or CCG unless doing benchmarking exercises). CSL will retain a maximum of 5 years of HES data. This will be on a rolling basis, whereby old data is destroyed as new data is received. Purpose 1 (patient pathways and variations by organisational/patient factors) To show aggregated patient pathways through the hospital system and provide an understanding of how patients are treated, and how treatment differs by key factors such as Trust, CCG or patient demographic (currently provided to NHS, medical suppliers and pharmaceutical companies) This is the most common type of analysis requested by CSL customers, allowing them to compare and quantify diagnosis and treatment patterns. This information is used by NHS organisations to identify where costs are being incurred and could be avoided, or where resources could be better focused to improve patient care or make scarce resources go further. It is also used to understand treatment pathways within hospitals, referral patterns and to help in system redesign. NHS organisations also use this data to review key data that they are obliged to monitor by the government. For instance, the level of surgical site infections, which CSL intends to provide free of charge to the NHS organisations in 2016. Type 1 example: As an example of data use for this purpose for a Type 1 organisation, CSL have been asked by an NHS Trust to help them quantify patients presenting with a combination of diagnoses that could indicate a specific undiagnosed condition. Type 2 example: NHS suppliers share this information with their NHS customers to help them identify potential opportunities for improving the effectiveness of treatment, or delivering cost reductions. These analyses also enable them to put together the cost-benefit analyses that are required as part of the process of gaining NICE approval or getting on formularies. For an example involving a Type 2 organisation, a medical supplier will use the data to provide an analysis of surgical site infections following particular types of procedure to NHS Trusts they supply. This will aid in the (Trust’s) financial evaluation of alternative treatment options in an area that is high on their priority list. For analysis in this area to be effective, CSL requests permission to hold a rolling 5 year period of data. This enables CSL to: • Define with some confidence (depending on the disease area) cohorts of patients with newly diagnosed conditions i.e. those that have not been admitted for the condition for a number of years • Analyse aggregated patient journeys over extended periods, to provide proxy’s for patient outcomes and identify relapses that may occur years after treatment Provide more substantial aggregated information on rare diseases, where numbers in any one year are too small to enable robust analysis. Purpose 2 (Benchmarking – other organisations and good practice) To compare actual treatments with best practice and NICE guidelines and to contrast Trusts and CCGs with each other. Both NHS and supplier organisations use this type of analysis to identify how trusts and CCGs are performing when benchmarked. This enables Type 1 NHS organisations to compare how they are performing with other similar organisations, and to identify areas where they are significantly different to their peers, or are divergent from NICE guidelines. This assists them to spot anomalies and recognise areas where procedures need to be reviewed. As an example of data use for this purpose for a Type 1 organisation, CSL have recently worked with a CCG to compare the frequency with which a particular procedure is undertaken in similar patient cohorts at the different Trusts they commission services from. A wider comparison was also conducted nationally to add context to the analysis. This has enabled the CCG to identify areas where pathway redesign is necessary and to open a dialogue with one of the Trusts they use to regarding the appropriateness of the level of procedures being carried out. NHS supplier organisations also use this information to plan initiatives or services to support the NHS or help them meet NICE guidance and reduce inequalities. For example with a Type 2 organisation, a supplier of specialist post-operative care equipment share analysis, including HES data, with their hospital customers. This helps these Trusts monitor their adherence to NICE guidelines in this area of shared clinical expertise. Purpose 3 (Identification, implementation, and monitoring of improvement plans) To quantify patient subsets to identify opportunities for cost savings or joint working initiatives and to monitor these initiatives once in place. This is often related to Purpose 2; having identified areas of development within a Trust or CCG, NHS and supplier organisations work together to improve a particular area of delivery within the NHS. The data CSL provide their customers is used as a trusted common source to identify and quantify the needs and opportunities for improvement for a given initiative and then to monitor progress over time. For example with a Type 1, data will be used to support an evidence based cost model to help NHS trusts and a life science company to work together to replace an existing invasive procedure with a new pill based treatment where clinically appropriate. As a further example for a Type 2 organisation, CSL provide data to a supplier of medical devices who share these data with their customers in NHS Trusts to help those Trusts meet NICE guidelines for post-operative patients.
Compufile Systems Ltd has helped deliver the following benefits to the English healthcare system: With CSL’s work with NHS Organisations • Enabled commissioners to make more informed decisions with regard to the care they provide their patients or in the evaluation of cost saving opportunities. For example, enabling users to identify avoidable hospital admissions and monitoring levels of hospital acquired infections. Analyses can be conducted by the client via ESPRIT, or by CSL staff on their behalf. The data can be used to quantify the number of patients in a particular cohort so that the cost of current treatment patterns can be estimated and compared to alternative methods or pathways of treatment. This, along with other evidence, helps commissioners to make decisions on whether to change the way patients are treated. E.g. In a recent study for a local CCG, CSL were able to demonstrate that their patients were receiving far more of a particular operation than would be expected, enabling them to review this with the Trusts delivering the care. • Identification to CCGs and Trusts of potential for efficiency gains in healthcare provision. For example, CSL have been able to identify instances where a particular Trust conducted multiple operations over multiple episodes rather than at once as was the norm in other Trusts. This was more expensive for the CCG and probably more inconvenient for the patient. With this data the CCG was able to investigate further with the Trust in question. • Evaluation of adherence to pathway guidelines and to create evidence based comparisons of patient journeys treated at different Trusts to help share best practice across healthcare providers. E.g. CSL have worked with a medical supplies provider to measure the adherence to NICE guidelines in patients undergoing anesthesia, to enable them to highlight where patient care could be improved. CSL have agreements in place with all clients that will require them to provide CSL with evidence of benefits to the health and social care system generated via their use of the data during 2016. In 2016, CSL intend to continue with some of the ongoing initiatives above, as well as providing some additional benefits: • Helping NHS suppliers (such as pharmaceutical and medical device companies) to identify joint working opportunities with hospital Trusts or CCGs to deliver cost savings or enhance patient care services. Unfortunately a recent example involving a first-to-market treatment for a rare form of cancer ceased before patient benefits were realised due to the therapy being withdrawn. However, CSL intend to work on similar examples in the future. • CSL have also been requested by a hospital Trust to help them quantify patients with a particular set of characteristics which could suggest an undiagnosed condition. CSL’s outputs will be used to quantify patients with certain combinations of diagnoses and demographics that physicians believe could be an indication that a true diagnosis has been missed. This information will help them to develop improved methodologies for identifying potentially undiagnosed or misdiagnosed patients, and ensure that previously undiagnosed patients receive appropriate care and treatment. Commericial Considerations: CSL is providing the NHS and those providing products and services to it with a toolset to enable them to analyse data, and in some cases consultancy to help them understand the results. CSL provides a basic service to the NHS free of charge and further services at a cost, subsidised through charging NHS Suppliers for these services.
All outputs from ESPRIT are shown as aggregated data (with small numbers suppressed in line with the HES Analysis Guide). It is not possible to see record level results. The data is provided to the user as cross-tabulations, charts and flow diagrams or reports. Each output is filtered to include the data relevant to the question being asked, and is then cross-tabulated by variables important to the analysis, such as hospital Trust or patient diagnosis to provide the data necessary to meet the objectives outlined above. ESPRIT does not provide healthcare professional level data to clients, to ensure that it cannot be used for targeting or direct marketing. Analyses are both conducted as a one-off or updated on an ongoing basis to monitor changes within care provision.
Use of HES data to populate the ESPRIT tool: The pseudonymised patient records are only held by CSL, and are imported into a secure database where they are organised for analysis within CSL’s own software (ESPRIT). No patient data are linked to other data sources at a patient level. Aggregated patient data are provided in ESPRIT alongside other practice level open data, including QoF and Deprivation statistics. ESPRIT sits on top of the record level data (under several layers of architecture security) and aggregates information in real time when a query is processed. This allows the data to be queried without the need to provide the user with access to the raw data, and only provides aggregated, suppressed results in line with the HES Analysis Guide. Use of ESPRIT tool by CSL CSL often conducts analyses for clients. CSL use the ESPRIT tool to conduct these analyses and provide the outputs to clients. Use of ESPRIT tool by 3rd parties CSL permit 3rd parties to access the ESPRIT tool to conduct their own analysis only under a controlled process. Before providing data to any organisation, CSL go through a multi-layered procedure to ensure the organisation and each user who will access the data is aware of the limited way the data may be used and the consequences of its misuse: • Organisations are provided with an overview document stating the limitations and regulations applying to the data’s use during the early stages of the engagement process, and prior to any commitments being made by CSL • Organisations sign a contract which passes on the pertinent terms and conditions from CSL’s framework agreement with the HSCIC. This ensures that data is not used for marketing purposes. • Organisations complete a purpose document, which requires them to state (with reference to CSL’s permitted purposes) how they will use the data, and the benefits they expect to deliver to the Health and Social Care System by doing so • This is reviewed by the Board at CSL, who ensure that the prospective customer’s stated purposes are subsets of CSL’s permitted purposes (both in word and spirit). Where they are not, the customer is informed and given the option to drop this purpose or walk away. • A further document is then completed by the prospect to state the subset of the data they require, the named users who will access it, and some relevant security details • Once this is in place, access is only granted following the completion of mandatory compliance training covering once more how the data must be used. Clients are reminded of the objectives for which the data may be used, and that data must be used to improve patient care and/or to support cost benefit analyses to assist commissioning and/or reduce treatment costs. This is conveyed via a message in ESPRIT each time it opens or footer on reports containing HES data. Each time a user opens CSL’s portal (ESPRIT) to view data, the login screen reminds them of the purposes of use permitted under this data sharing agreement, and that by logging into the system they agree to this restriction. Contracts with clients enable CSL to enforce these terms and terminate the contract with immediate effect if they are not adhered to.