Update from the The Health and Social Care Information Centre
Chris Dew Head of Clinical indicator Development at The Health and Social Care Information Centre delivered a presentation on Estimating the risk of mortality in hospital patients. Chris discussed
this estimation underpins the calculation of the Summary Hospital level Mortality Indicator (SHMI), which is published quarterly by the HSCIC and is used to identify whether the number of deaths which have occurred among hospital patients is more or less than expected.
using this information to reduce mortality rates
Chris started his presentation by explaining that the HSCIC is the National provider of information, data and IT systems for commissioners, analysts and clinicians in health and social care. The Clinical indicators IT programme looks at developing methodologies, technical construction, assurance and dissemination.
Chris then focussed on; Why SHMI (Summary Hospital level Mortality Indicator)? Explaining that it’s a computer model which calculates a ratio of observed deaths against expected deaths, so more meaningful than just looking at the number of deaths. Established national baseline and compares Trusts against that. The calculation of risk – is adjusted based on different factors. E.g. underlying factors, age, pre-existing conditions.
There was lots of discussion in the room about difficulty with the data and with how its coded – it can only look at data in certain ways, it doesn’t give contextual view. Doctors find this difficult because its ambiguous – when you start to look at individuals it becomes difficult.
Chris suggested it should be looked at as a ‘smoke signal’ that there may be problems/failings – SHMI must be used in the right, pragmatic way – trigger further investigation.
Still have sensible screening process to ‘weed out’ those deaths which were ‘inevitable’ based on patient’s condition upon admission – i.e. when they were admitted did you ‘expect’ them to die? If so they should be excluded from possible ‘preventable’ category.
Helpful to highlight periods to investigate, e.g. if there was a ‘spike’ in number of deaths vs national level, you might want to investigate.
“It is NOT a league table, a direct measure/comparison, or a definitive judgement”
Developments – “weekend mortality” has been in media, they are looking at this at the moment.
Chris Dew’s full presentation is available to download at the end of this page
Abstract of Chris’ presentation:
Chris’s presentation will describe how HSCIC estimate the risk of mortality for hospital patients to produce the Summary Hospital-level Mortality Indicator (SHMI). This estimation underpins the calculation of SHMI, which is published quarterly by the HSCIC and is used to identify whether the number of deaths which have occurred among hospital patients is more or less than expected.
Chris Dew’s Biography:
Chris is responsible for the development and delivery of a variety of products including indicators underpinning the NHS Outcomes Framework and the CCG Outcomes Indicator Set and the Summary Hospital-level Mortality Indicator.
Chris’s team are also responsible for the Indicator Assurance Service. This was given statutory footing when the Health and Social Care Act (2013) tasked the HSCIC to establish procedures for the assessment of a “quality” indicator; and arrange for each quality assured indicator to be published in a national library and periodically reviewed. Chris is Deputy Chair of the Methodology Review Group which forms part of this service, offering independent peer review of the methods proposed for new indicators.
Since graduating from the University of Durham with a BSc in Mathematics Chris has spent 13 years in the Government Statistics Service for the Department for Work and Pensions and the Health & Social Care Information Centre.
Outside of work Chris is a keen football fan and traveller and has often combined the two to see Sunderland lose in a variety of places…
Also of interest:
Delivering a 7 Day Health Service
Clinical Audit Masterclass
National PROMs Summit 2015
Data Quality and Clinical Coding for Improvement