NHS Digital Data Release Register - reformatted

Crab Clinical Informatics

Project 1 — DARS-NIC-351722-W7D4N

Opt outs honoured: N

Sensitive: Sensitive, and Non Sensitive

When: 2016/04 (or before) — 2018/05.

Repeats: Ongoing

Legal basis: Health and Social Care Act 2012

Categories: Anonymised - ICO code compliant

Datasets:

  • Hospital Episode Statistics Admitted Patient Care
  • Hospital Episode Statistics Critical Care

Benefits:

The CRAB system was deployed nationally by the Keogh review of 14 NHS Hospitals, and was instrumental in providing key lines of enquiry to investigating teams. It is now being trialled for ongoing use by the CQC. The benefit of CRAB to clinicians, managers and Board members at hospital Trusts includes the ability to: - understand quality of care on an ongoing basis, and to set in place appropriate governance and monitoring systems. - receive early warning of problems and monitor trends for deterioration in practice quality - rapidly investigate and interpret SHMI results (and other mortality data such as HSMR), to understand root cause and underlying patterns in relation to mortality, in accordance with DoH policy guidance that provider organisations should have their own localised clinical dashboards for this purpose. - monitor quality beyond basic mortality: assessing morbidity and avoidable harm, as a fundamental move towards continuous quality improvement - generate appraisal documentation for clinical staff - improve accuracy of coding - understand the clinical risks (or mortality and complications) for each individual patient and have frank and open discussions with them. These benefits can also be aggregated to a national level, in so far as the DoH and regulatory bodies (Monitor and CQC) can use the information to drill into more detailed analysis of organisational performance where appropriate, particularly in relation to morbidity and avoidable harm - as is envisaged to support CQC inspection activity and quality baselining by Monitor. In relation to the interactive databases, the reporting available is to individual patient level, enabling organisations to conduct detailed audits and investigations where necessary (e.g. critical incident & SUI reporting). However, in any given case, the raw data itself is not accessed, and in relation to the reports, these are only accessible to the organisation and staff concerned, being hosted in a dedicated, encrypted environment and subject to IG-compliant processes for log-in and individualised permissions. As indicated in relation to C-CI current licence for historical HES, C-CI’s data hosting partners, L2S2 Ltd, are IG Toolkit Level 3, and ISO27001 certified.

Outputs:

Analysis of clinical outcomes, comprising: - clinically risk adjusted surgical mortality and morbidity for the case-mix of patients treated. - review of avoidable harm events across surgical, medical and nursing care The data detailed will be used to: - provide overview reports to NHS bodies (national and local); - create interactive databases for those bodies to interrogate the reporting interactively; and – Potentially contribute to validation of future changes to CRAB algorithms (which are reviewed annually against a proprietary international dataset to accommodate any changes in clinical practice globally). Reviews may be one-off (commissioned either locally or nationally), or where an ongoing license with an organisation has been agreed, the above analysis will be available both from the live interactive database described above for the NHS Client’s use, and scheduled monthly reports will also be generated to track immediate trends by way of internal clinical governance and service management. Reviews and scheduled reports of acute care and performance of an organisation are for that organisation only, unless being reviewed for a national body such as the CQC (hospital inspection team) or Monitor, with timescales determined by the executives. Further details below.

Processing:

In so far as Clinical Informatics Limited create “interactive databases for those client NHS organisations to interrogate”, this involves setting up bespoke CRAB databases for each organisation that purchases the software: these databases allow for the client to pull up standard CRAB reports, and also, to a certain degree, to make bespoke queries. However, in all cases, the database they interrogate contains only data relating to that NHS organisation (and it is already data that they have submitted centrally), and the access is only to CRAB analysis and reports, not to the raw data. Data is loaded onto CRAB server hosted at L2S2 where individual database are created for individual NHS trusts data only. Here they can be accessed over N3 by C-Ci staff or NHS employees (please refer to data flow diagram) The HES data will not be linked to any other data other than publicly available data, or (as anonymised output) to other data relating to the Trust in fulfilment of the purpose outlined within this application. Data can only be viewed over the N3 network, and customers can only view the data and recall CRAB reports. It is not possible to download the raw data. No record level data will be stored outside L2S2. All outputs are aggregated with small number suppression in line with the HES Analysis Guide.

Objectives:

Crab Clinical Informatics Limited (C-Ci) is responsible for developing and marketing CRAB (Copeland’s Risk Adjusted Barometer). Designed by the former National Director for Clinical Audit, CRAB is a web-based tool to evaluate quality and outcomes in a way which accurately reflects the clinical profile of patients treated. This is designed to provide a granular local dashboard to help Trusts interpret mortality analysis (HSMR/SHMI) and understand safety in relation to avoidable harm, morbidity and areas for improvement. CRAB comprises two modules - Surgical and Medical. Reporting is designed to underpin the clinical governance strategy for individual and organisational practice. The focus is on local service line management and appraisal. No PID will be available or shared through CRAB Reviews. Trusts own pseudonymised data is available for their own internal reviews, investigations and appraisals, and not available to anyone outside the organisation other than C-Ci staff and agents (and national authorities upon request as indicated elsewhere in this application) under the terms of this agreement and strict ISA terms with the client Trust. To clarify, The reviews C-Ci will be producing using central HES will not be making patient level pseudo data available to anyone other than C-Ci staff for the purposes of compiling aggregate analysis. C-Ci’s central HES reviews are principally intended either for national bodies, or as initial baseline assessments for NHS Trusts which request them. Some NHS Trusts may wish to progress to a fully installed CRAB system. No patient level data provided under this agreement is provided to those Trusts. Those Trusts will enter in to a separate agreement with CRAB for a flow of local SUS data, but that is outside this agreement, and the data provided under that local flow is not used in conjunction with any record level data under this agreement. Agents and national authorities : National authorities on behalf of which C-Ci may generate reports comprise specifically: CQC, Monitor, Department of Health. C-Ci in addition has worked on a case-by case basis with the following commercial agents: PwC (as part of the Keogh Reviews and any follow up thereto), & Oliver Wyman, for specific quality improvement projects with the engagement of the client NHS Trusts. In all cases, neither the agents nor the authorities have access to the raw HES data, only the CRAB analysis. - CRAB in Surgery generates case-mix adjusted outcomes using POSSUM (all variants) – a respected audit methodology, recommended by the RCS. - CRAB Medical produces reports detailing trigger events which are indicative of iatrogenic harm. These can be grouped to understand overall quality of care, and potential deficiencies in process. Trigger events are triangulated from the published UK version of the Global Trigger Tool, a recognised approach for quality improvement in use widely as a manual sampling methodology across the NHS. The HES dataset is being obtained to: - assist in objective benchmarking of organisations, - expedite quality analysis potentially on behalf of the CQC (in process - to be confirmed) - replay CRAB analysis and reporting to individual client NHS organisations. The analysis provided to the Keogh review team was based on a bulk extract of HES & ONS data supplied by the HSCIC for the 14 hospitals concerned. A similar exercise was also conducted using HES data on behalf of Monitor, and again as a research project investigating emergency laparotomy practice with Royal United Bath Hospital NHS Trust. C-Ci Ltd. therefore has an established record of receiving and handling data from the HSCIC. With the extension of this system following the Keogh Review and the newly defined CQC inspectorate, as well as to make the analysis more widely available to Trusts individually, this application is to seek a licence of HSCIC data on an ongoing basis. In order to produce an accurate and reliable quality review of an organisations care they are delivering at that point in time, C-CI require the most current available data to make a worthwhile analysis. If non-current data is used then care practices and standards may have changed since C-CI produced the review on the data C-CI have available. Organisations may already have provided additional resources to where remedial measures were required, so C-CI wouldn’t want to suggest they add to that effort if they are not required. On the other hand, areas of poor care identified may have deteriorated even further, and require more resource than originally suggested. The HES data set is used to highlight any areas of care where standards require improvement, or vice versa where good care practices can be rewarded and lessons learnt from where good care exists. C-CI want to provide the most accurate account possible of the care being delivered by an organisation, so the organisation can see the whole picture and make decisions based on their current data performance. The same argument also holds true for our work for the CQC, where the intention is to provide targeted, key lines of enquiry for on-site inspection teams, and therefore the more current the data, the more salient the lines of enquiry will be. C-Ci products and information is used to examine an organisation’s clinical performance. C-Ci then provides expert remedial advice if any areas of concern are highlighted. The same analysis may also be commissioned by other NHS/Government bodies (to date, CQC, Monitor, the DH and NHS England) as previously done for the Keogh Review and also ongoing CQC inspections, and in that event reports are produced for the same purpose and will not be shared with any other organisation beyond the commissioning organisation and the target NHS Trust. C-Ci does work strategically with healthcare management consultants (previously PwC for the Keogh reviews, and currently exclusively Oliver Wyman), for the purposes of enabling Trusts strategically and operationally to make the most of the CRAB system. This is through offering a supported programme to implement the organisational changes necessary (both cultural and systemic) to achieve improvements in quality identified by CRAB analysis. This has been flagged in previous documentation. CRAB analysis in the form of static reports may therefore be made available to Oliver Wyman in the context of initiating collaborative improvement projects with the NHS Trusts concerned, but under no circumstances would Oliver Wyman be given access to the raw data, nor is the live CRAB database shared with them. Such analysis provided to Oliver Wyman is in aggregate (trend lines/bulk period analyses) form only – small numbers are suppressed and there is no reference to record-level data. In terms of customer base, presently approximately 95% are NHS providers/commissioners, and 5% are national bodies. This excludes the single commercial sector organisation using CRAB reports as outlined above.