Summary Hospital-level Mortality Indicator (SHMI) Statement of Administrative Sources: Annex B supporting metadata

Organisation

NHS England

Enquiries contact 

Contact the NHS England contact centre.

Responsible statistician

Stephanie Gebert

Section: Clinical Outcomes and Indicators

Summary Hospital-level Mortality Indicator (SHMI) and supporting contextual indicators

The SHMI reports on mortality at trust level across the NHS in England using a standard and transparent methodology.  This indicator is produced and published monthly as an Accredited Official Statistic by NHS England.

The SHMI is the ratio between the actual number of patients who die following hospitalisation at the trust and the number that would be expected to die on the basis of average England figures, given the characteristics of the patients treated there. 

Name

Hospital Episode Statistics (HES) Admitted Patient Care (APC) data linked with death registrations data from the Office for National Statistics (ONS).

Name of the organisation responsible

HES APC data: NHS trusts are responsible for submitting data to the Secondary Uses Service (SUS).  The data are then processed by NHS England to create HES data.

Death registrations data: Office for National Statistics (ONS).

Purpose of the administrative source

HES APC data: Data are submitted to SUS to enable healthcare providers to be paid for the care they deliver.  The data are then processed by NHS England to create HES for the purposes of statistical analysis.

Death registrations data: Official register of all deaths.

Unit of inquiry

HES APC data: Finished consultant episodes (a continuous period of admitted patient care under 1 consultant within 1 healthcare provider).

Death registrations data: Registered deaths.

Intended coverage

HES APC data: All NHS funded admissions at English hospitals.

Death registrations data: All deaths in England and Wales.

Actual coverage

HES APC data: Episodes submitted by healthcare providers to SUS.

Death registrations data: All registered deaths in England and Wales

Geographical coverage

HES APC data: England

Death registrations data: England and Wales

Lowest level of geographical coverage

HES APC data: Hospital provider site

Death registrations data: Local authority

Extent to which statistical end-producers can influence system?

HES APC data: The Data Alliance Partnership Board (DAPB) controls the information that is collected from hospital administration systems.

Death registrations data: The death registrations data are covered by the Statistics and Registration Service Act 2007.  Sections 39(4) and S42(4) apply.

Data definitions used

HES APC data: HES data dictionaries are available.

Death registrations data: Further information is available.

Classification systems used

HES APC data: Diagnoses are coded using the 10th revision of the World Health Organization's International Classification of Diseases (ICD-10).

Procedures and interventions are coded using OPCS codes. 

Death registrations data: Cause of death is coded using ICD-10 see above. 

Data collection process

1. Periodicity/Timing

Provisional monthly data and final annual data are used in the calculation of the SHMI (for both HES APC and death registrations data).

2. Validation process

HES APC data: Data validation checks are carried out when trusts submit data to SUS. In addition, further data validation and cleaning is carried out as part of the production of the HES dataset.  The HES Data Quality team discuss data quality issues with the information leads in organisations who have submitted the data and data quality notes are available. 

Death registrations data: Routine data validation checks are carried out by the registrar when a death is registered. Further checks (for example to identify duplicate records and to check for consistency between dates of birth, death and registration) are carried out when the data are received by ONS and suspect records are referred back to register offices. Full details of the quality assurance process are available on the ONS website. 

Access arrangements

HES APC data: HES data are not obtained directly from patients but from hospital provider systems via SUS. It is not feasible to obtain explicit consent from patients for the multitude of uses to which HES data can be put. NHS Digital has statutory powers through the Health and Social Care Act 2012 to receive, store and process patient identifiable data. Access to this level of data by another organisation is controlled by NHS Digital’s Data Access Request Service (DARS).  Individuals can choose to opt-out of their confidential patient information being shared or used for any purpose beyond providing care.  Further information on the different types of opt-outs is available. 

Death registrations data: Access to mortality data is governed by the Statistics and Registration Service Act 2007. Sections 39(4) and S42(4) apply.

Dissemination procedures

Not applicable 

Publication procedure

Not applicable 

Timing/periodicity of public release

Not applicable 

Nature of changes to system that could impact on statistics

Any change to the data could affect the statistics.  These are considered on an individual basis and are explained and footnoted in publications where appropriate.

HES APC data: System changes are implemented using Information Standards Notices (ISNs).

Death registrations data: All changes and updates are detailed in the documentation released by ONS alongside the data. National consultations and expert peer reviews are all conducted by ONS.

Validation procedures

Data processing and statistical modelling are carried out according to the published methodology specification document using peer reviewed processes.  Activity counts are compared to those for the previous release and any large changes are investigated.  The HES data quality notes are also reviewed and details of any data quality issues impacting on the publication are highlighted in the outputs and in the background quality report. 

On a quarterly basis, trusts are provided with pre-release access to their own SHMI data for the purpose of quality assurance via NHS Digital’s Clinical Indicator Preview tool.  Each preview period allows 5 working days for trusts to carry out this quality assurance and to raise any issues or questions directly with the production team.  If trusts raise data quality concerns during the preview period, the impact of the issue is investigated, and this is then documented in this background quality report where necessary.   

Quality assessment

A background quality report is produced as part of each monthly publication.

Periodicity of release

Monthly

Potential impact of changes to the source on the statistics produced

Substantive changes to the source data will result in breaks in the data series for example new diagnosis codes and changes in geographical boundaries.

Procedures for mitigating such discontinuities

Discontinuities are assessed on an individual basis and are explained and footnoted in publications where appropriate.