NHS Digital Data Release Register - reformatted

Birmingham Children's Hospital NHS Foundation Trust

Project 1 — DARS-NIC-11544-S1L0R

Opt outs honoured: N

Sensitive: Non Sensitive

When: 2016/09 — 2018/05.

Repeats: Ongoing

Legal basis: Health and Social Care Act 2012

Categories: Anonymised - ICO code compliant

Datasets:

  • HES Data Interrogation System
  • Hospital Episode Statistics Admitted Patient Care
  • Hospital Episode Statistics Accident and Emergency
  • Hospital Episode Statistics Outpatients

Benefits:

Pilot access has been granted to HDIS through 15/16. During this time, BCH have been able to achieve many benefits which would not be possible without continued access to the HDIS database. For example, summary information has been extracted from HDIS to inform a forecasting model that provides a forecast of future surgical activity within the region forwards to 2021, with an onward look to 2030. This is being used to design future services, ensuring are sufficiently productive and well placed to deliver against demand in future years. BCH have on-going project which is exploring benchmarking possibilities against peer comparators to describe variations in paediatric care. BCH uses a programme management approach that understands benefits as the measurable improvements resulting from a set of outcomes that are enabled by the organisations changing capabilities. The capabilities expected from HDIS is to give the organisations are; • Access to national datasets on inpatient, outpatient and emergency activity that enables the trust to take a population (rather than point) view of healthcare delivery and performance • The ability to create custom queries that filter by demographic and clinical factors to extract information on incidence, care pathway and outcomes • Ability to produce healthcare needs assessment on targeted conditions or patient groups • Ability to produce healthcare needs assessment for wider paediatric populations. The outcomes this will enable include; • Improvements in planning across the health economy for children and young people • Higher priority to health inequalities that affect children and young people • Better planning for the population level needs of children and young people • Reduction in variation of healthcare experience between different groups of the population • Improved forecasting of demand on services, and a more responsive healthcare system Benefits for these outcomes fall in the short and long term. For example, HDIS will support planning of the future configuration of paediatric services within the region, but this is unlikely to be fully realized until new estate becomes available in 2022. However, short term measurable benefits also exist; • Report on variation in emergency care will trigger initiation of a project in collaboration with commissioners to explore variation in asthma care experience, particularly exploring the artificial barriers that primary/secondary care introduce into the system (now completed September 2015) • Reduction in variations of length of stay between BCH and other organisations for key areas, including elective surgical patients and oncology treatments. (now completed October 2015) • A description of the emergency attendance behaviours around paediatrics that will inform the child health network in Birmingham for its planning of a response to the Keogh review. (due September 2016) • A healthcare needs assessment of changes in paediatric attendance patterns through winter to inform contribution of organisation to the sustainability and transformation planning needs (due August 2016) • Upstream study looking at maternal presentations and anticipating impact on births and neonatal flows (Due July 2016) Together, it is intended that the benefits of use of the HDIS system by BCH will be an outcome for children, young people, families and the health and social care system of a well planned system of paediatric care, that delivers excellent care, and continual improvement. It is intended that HDIS will support making this care accessible, adaptive and resilient to the levels of population need within local, regional and national populations.

Outputs:

All outputs are produced to a standard specification, which includes aggregation of small number data in line with the HES analysis guide. In some cases, this includes withholding fields where they provide information about other parts of the final information set. HDIS is a flexible system which allows for the ongoing production of healthcare needs assessments, benchmarking of services, and support to the development of regional and national strategy towards the prevention and treatment of paediatric diseases. Specific use cases identified are: 1. Exploration of regional asthma emergency admission activity for 0-14 year old children living in the West Midlands metropolitan county. National data demonstrates that several areas in the West Midlands have higher than expected rates of paediatric admissions. BCH has undertaken several initiatives to improve the care pathway and discharge of patients with this condition. However, the variation remains. Analysis of internal hospital datasets cannot show us the picture across the region, because it only captures patients admitted to a single site. Using HDIS, it has been possible to audit services and engaged with commissioners to explore variation in referrals between general practices, and across areas of the city. It is intended to use HDIS to filter activity to specific lengths of stay, ages and diagnostic categories, but being able to present a population need, rather than a provider delivered view. It is anticipated that this information is presented in a standardised rate form, grouped to remove small numbers, and with smaller practices being clustered to again protect the identity of small patient groups. 2. Retrospective analysis of paediatric emergency department attendances and conversion rates through winter. Paediatric services, in line with adult services, have had substantial pressure through the last winter. BCH has had several periods of special cause variation in admission rates. HDIS will enable us to correlate whether this activity represented true increases in activity, or shifts in activity from other providers. This will inform future planning, enabling to identify whether prevention, capacity or health seeking behaviour is the root cause, thus ensuring that it can demonstrate better preparedness in future periods of greater pressure. 3. Healthcare needs analysis of surgical activity within the West Midlands. Due to changes in clinical governance thresholds, retirement of senior surgeons and re-configurations of several services within the region, demand on surgical pathways has proved volatile and hard to forecast. It is intended to use HDIS to demonstrate how surgical flows are changing, and whether there are particular geographies where there are substantial shifts in the management of children and young people. The intelligence produced through the system will enable and ensure a balance is used for services across the region to meet demand, and are able to have constructive discussions within the health economy on how best to meet the population health need. 4. BCH has committed to extend its role as an advocate for better paediatric public health, health and social care for children, young people and families. HDIS enables to act within the wider health and social care system to challenge on health inequalities for children. It is intended to use the system to look at variations in attendance, conversion and treatment experience for children and young people, and use this to make the case for necessary system wide improvements. For example,recently using the HDIS system to examine attendance rates by social deprivation and ethnicity, by age group and emergency department. This allowed to demonstrate different local behaviours, and that the apparent geographical preference suggested by internal data was not validated when examined at a population level The majority of aggregate information (already compliant with HSCIC small numbers standards etc.) is only shared with specific individuals relevant to a project or workstream. Occasionally, specific relevant and appropriately grouped outputs with be shared with other organisations. Certain projects will produce custom indicators of performance. These are used for the purposes of quality improvement, either internally, or in discussion with other organisations. All outputs will be aggregate with small numbers supressed in line with the HES analysis guide. Data accessed via the HDIS database will not be used for any commercial purposes. No publishing of the findings of HDIS queries are made through journals. Academic output is not a requirement for funding of HDIS within the organisation.

Processing:

The application is for online access to the HDIS system. BCH NHS FT have a medical consultant in public health who is sole user authorised to use the HDIS system, whose skillset includes SQL query writing and advanced statistics, as well as undertaking the complex joins required to process data online. A multi-step processes is used within the HDIS virtual environment to ensure that the majority of processing is completed within the HSCIC virtual environment. Linkages of HSCIC datasets only occurs within this single virtual environment. Patient level data is never extracted from the raw data files within the system, nor are small number values. Once the query within the virtual machine is complete, extract data is put into an excel spreadsheet, keeping an audit trail of the query undertaken and the summary extracted for review by our internal governance team, as well as external bodies such as the HSCIC. Once extracted from the system, data is stored in a secure NHS network environment with user access control to the files. Local processing is used to format or visualise the data (e.g. bar charts). A final manual audit check is undertaken on the data to confirm that it is compliant with HSCIC standards, for example confirming it is not possible to derive obfuscated values from column totals etc. The data is never transferred into any cloud based systems, nor analysed outside of the local environment. For certain analyses, data is linked locally to commonly available summary datasets – for example, using the LSOA field within the HDIS dataset to link to a specific deprivation indicator. This is only applied at the summary data level. The data is not processed by any additional individuals, and is only released from the sole user once processing of the information is complete.

Objectives:

Birmingham Children’s Hospital is a large provider of local, regional and national care for Children, including several nationally commissioned services. It is intended to process the data available through HDIS to support three activities; • Challenging practice, both at the hospital and across the local economy, through the production of benchmarks for length of stay and outcomes, conversion rates from emergency department attendance, and exploring variation in care over time • Contributing to the whole health economy planning processes through strategic analysis of patterns of paediatric activity, including the impact of re-configurations occurring at other sites which are decreasing their contribution to the provision of paediatric care, and implementation of the Keogh review recommendations. • Exploring the strategic future possibilities for Birmingham Children’s Hospital, in particular describing the likely levels of population need as it is planned to build a new facility for Birmingham in 2023/4. To make sure that this is balanced against the needs of the local, regional and national populations that is served. Given the specialist nature of the activity that are delivered, and the wide geographies that patients travel from, HDIS gives the unique ability to design queries that can be appropriately filtered for demographic and clinical care factors, allowing BCH NHS FT to produce standardised information. In particular, it allows querying of subgroups of age/residence at a granularity not available in national datasets. This is of particular importance in a context focussed only on treating children. The majority of the intelligence that are produced using the HDIS system is used internally, in discussion with senior leaders and clinicians at the trust. The summary information produced from HDIS is also used in discussions with other organisations, including commissioners. Before data is presented or shared, it is audited as fully compliant with HSCIC recommendations on small numbers, diagnoses, and geographies. The trust uses standard filter rules in HDIS to remove inappropriate data, and all data is subsequently audited against a template before use. Occasionally, data will be used from the HDIS system in reports on paediatric health that are released into the public domain. Again, these will be audited for compliance against HSCIC standards for protecting patient anonymity. No data included in these reports is at a granularity that would not be available through a freedom of information request route. Raw data is never extracted from HDIS, and no data below summary granularity is shared with partners outside of these terms. All outputs produced will be aggregate with small numbers supressed in line with the HES analysis guide. Data will only be used for purposes relating to the provision of healthcare or the promotion of health in line with the requirements of the Health and Social Care Act 2012 as amended by the Care Act 2014.