Learning from deaths Regional Mortality and Mortality Review Thursday 25 th May 2017 Durham Centre, Durham Mr Tony Roberts 1
Deputy Director (Clinical Effectiveness) South Tees Hospitals NHS Foundation Trust www.southtees.nhs.uk Patient Safety Collaborative Programme Lead, Academic Health Science Network North East and North Cumbria, (AHSN NENC), www.ahsn-nenc.org.uk Deputy Director North East Quality Observatory Service (NEQOS) www.neqos.nhs.uk Member of the Q Initiative www.q.health.org.uk 2
Overview brief overview of the CQC review into investigations of deaths in NHS Trusts and the new regulations for Trusts to estimate avoidable deaths findings and learning from our regional mortality review programme implementing the National Quality Board National Guidance on Learning from Deaths: Identifying, Reporting, Investigating and Learning from Deaths in Care thoughts so far on how we might meet these requirements 3
New requirements from April 2017 http://www.cqc.org.uk/content/learning-candour-and-accountability https://www.gov.uk/government/speeches/cqc-review-of-deaths-of-nhs-patients 4
Seven Recommendations Recommendation 7: Provider organisations and commissioners must work together to review and improve their local approach following the death of people receiving care from their services. Provider boards should ensure that national guidance is implemented at a local level, so that deaths are identified, screened and investigated, when appropriate and that learning from deaths is shared and acted on. Emphasis must be given to engaging families and carers. 5
National Quality Board March 2017 NHS Anytown Foundation Trust: Learning from Deaths Dashboard - September 2017-18 Description: The suggested dashboard is a tool to aid the systematic recording of deaths and learning from care provided by NHS Trusts. Trusts are encouraged to use this to record relevant incidents of mortality, number of deaths reviewed and cases from which lessons can be learnt to improve care. Summary of total number of deaths and total number of cases reviewed under the Structured Judgement Review Methodology Total Number of Deaths, Deaths Reviewed and Deaths Deemed Avoidable (does not include patients with identified learning disabilities) Total Number of Deaths in Scope Total Deaths Reviewed This Month This Month This Month Last Month Last Month Last Month 454 339 14 523 298 20 This Quarter (QTD) Last Quarter This Quarter (QTD) Last Quarter This Quarter (QTD) Total Number of deaths considered to have been potentially avoidable (RCP<=3) Last Quarter 1436 939 50 1509 1053 54 This Year (YTD) This Year (YTD) This Year (YTD) Last Year Last Year Last Year 6069 3991 227 0 0 0 Time Series: Start date 2017-18 Q1 End date 2018-19 Q2 Mortality over time, total deaths reviewed and deaths considered to have been potentially avoidable (Note: Changes in recording or review practice may make comparison over time invalid) 1800 1600 1400 1200 1000 800 600 400 200 0 Q1 2017-18 Q2 Q3 Q4 Q1 2018-19 Q2 Total deaths Deaths reviewed Deaths considered likely to have been avoidable Total Deaths Reviewed by RCP Methodology Score Score 5 Score 1 Score 2 Score 3 Score 4 Score 6 Definitely avoidable Strong evidence of avoidability Probably avoidable (more than 50:50) Probably avoidable but not very likely Definitely Slight evidence of avoidability not avoidable This Month 0 0.0% This Month 4 1.2% This Month 10 7 2.9% This Month 33 9.7% This Month 65 19.2% This Month 227 67.0% This Quarter (QTD) 5 0.5% This Quarter (QTD) 14 1.5% This Quarter (QTD) 31 3.3% This Quarter (QTD) 90 9.6% This Quarter (QTD) 178 19.0% This Quarter (QTD 621 66.1% This Year (YTD) 30 0.8% This Year (YTD) 65 1.6% This Year (YTD) 132 3.3% This Year (YTD) 378 9.5% This Year (YTD) 754 18.9% This Year (YTD) 2632 65.9% Summary of total number of learning disability deaths and total number reviewed under the LeDeR methodology Total Number of Deaths, Deaths Reviewed and Deaths Deemed Avoidable for patients with identified learning disabilities Total Number of Deaths in scope Total Deaths Reviewed Through the LeDeR Methodology (or equivalent) Total Number of deaths considered to have been potentially avoidable This Month This Month This Month Last Month Last Month Last Month 10 10 2 2 2 0 This Quarter (QTD) This Quarter (QTD) This Quarter (QTD) Last Quarter Last Quarter Last Quarter 16 16 3 24 24 4 This Year (YTD) This Year (YTD) This Year (YTD) Last Year Last Year Last Year 75 75 19 0 0 0 Time Series: Start date 2017-18 Q1 End date 2018-19 Q1 Mortality over time, total deaths reviewed and deaths considered to have been potentially avoidable (Note: Changes in recording or review practice may make comparison over time invalid) 35 30 25 20 15 10 5 0 Q1 2017-18 Q2 Q3 Q4 Q1 2018-19 Total deaths Deaths reviewed Deaths considered likely to have been avoidable https://www.england.nhs.uk/wp-content/uploads/2017/03/nqb-national-guidance-learning-from-deaths.pdf https://www.england.nhs.uk/ wpcontent/uploads/2017/03/nqb -learning-from-deathsdashboard.xlsx 6
From April 2017 the NQB National Guidance on Learning from Deaths require: Acute, mental health and community NHS Trusts and Foundation Trusts should ensure their governance arrangements and processes include, facilitate and give due focus to the review, investigation and reporting of deaths, including those deaths that are determined more likely than not to have resulted from problems in care. Trusts should also ensure that they share and act upon any learning derived from these processes. Non Executive Director and Executive Director Board responsibility Enhance skills and training A clear policy for engagement with bereaved families and carers Each Trust should publish an updated policy by September 2017 on how it responds to, and learns from, deaths of patients who die under its management and care. 7
From March 31 2017 the boards of all NHS Trusts and Foundation Trusts will be required to: The policy should include how its process respond to the death of an individual with: - a learning disability - a mental health need - an infant or child death - a still birth or maternal death Undertaking case record reviews: selection of cases, method and reporting From April 2017, Trusts will be required to collect and publish on a quarterly basis specified information on deaths. This should be through a paper and an agenda item to a public Board meeting in each quarter to set out the Trust s policy and approach (by the end of Q2) and publication of the data and learning points (from Q3 onwards) Changes to the Quality Accounts regulations will require that the data providers publish be summarised in Quality Accounts from June 2018 8
From March 31 2017 the boards of all NHS Trusts and Foundation Trusts will be required to: In 2017-18, further developments will include: The Care Quality Commission will strengthen its assessment of providers learning from deaths NHS England, led by the Chief Nursing Officer, will develop guidance for bereaved families and carers. Acute Trusts will receive training to use the Royal College of Physicians Structured Judgement Review case note methodology NHS Digital is assessing how to facilitate the development of provider systems and processes The Department of Health is exploring proposals to improve the way complaints involving serious incidents are handled 9
Medical Examiner reforms: In at least two Trusts in England, every death in hospital is discussed with a Medical Examiner; first to decide whether the coroner needs to be informed, then to agree how best to formulate the cause of death for the death certificate The medical examiner also examines the medical record, at least in relation to the terminal admission, and speaks to a member of the bereaved family, usually by telephone. The family member is offered an explanation of what is written on the death certificate as the cause of death, and is asked whether any aspect of healthcare could have been done better. If the ME views the body of the decease they can complete the confirmatory (Part 2) medical certificate for cremation if needed The service is funded from the fees. If any concerns are identified at any stage the case can be identified for more detailed review or investigation 10
National MCRRP for acute hospitals 11
Grading Deaths or Care Expected v unexpected death: A death can be said to be unexpected if: Avoidability Scale Descriptor The patient died of an unexpected illness not suspected by the managing team The diagnosis was suspected and the patient was treated but died despite not having bad prognostic features associated with that diagnosis The diagnosis was suspected and the patient was treated, however the treatment was sub-optimal. 1 Definitely not avoidable 2 Slight evidence of avoidability 3 Possibly avoidable (less than 50:50) 4 Probably avoidable (more than 50:50) 5 Strong evidence of avoidability 6 Definitely avoidable NCEPOD grade Descriptor PRISM Quality Scale 1 Good practice a Excellent 2 Room for improvement in clinical care b Good 3 Room for improvement in organisational care c Adequate 4 Room for improvement in clinical and organisational care d Poor 5 Less than satisfactory e Very poor Considering all that you know about this patient's admission, how would you rate the OVERALL quality of healthcare received by the patient from this trust? This question recognises that a problem in care can occur against a backdrop of overall good quality of care and the converse, a patient may experience poor overall quality of care w ithout obvious harm. For this question, do not consider healthcare prior to the admission that ended in the patient's death. 12
A brief history in the North East JUL 2012 PRISM 1 study Estimated avoidable mortality 5.2% 13
A brief history in the North East JUL 2012 PRISM 1 study Estimated avoidable mortality 5.2% NE Regional Mortality group collaborate to adopt locally adapted PRISM methods 14
A brief history in the North East JUL 2015 PRISM 2 study Estimated avoidable mortality 3.0% 3.6% 15
Mortality Review South Tees Centralised review system Commenced October 2013 2786 REVIEWS 16
Mortality Review South Tees Centralised review system In the last 12 months 676 REVIEWS 0.2% >50/50 preventable 2.7% some preventability 16% NCEPOD room for improvement 86% excellent, good 12% adequate 2% poor 17
Mortality Review South Tees Centralised review system Key Lessons Documentation Delays and problems with end of life care and DNACPR Issues with tertiary referral or repatriation 18
Mortality Review South Tees Centralised review system Team Based on 2 consultants Each 1 PA / week Administrator and In-house database Support from Nurses & Other Specialties 19
Mortality Review South Tees CASES: RANDOM 25% of all deaths And selected groups of patients Patients with any incident or complaint recorded Following surgery or a procedure 4 or more medical triggers Adults aged <50 Learning disabilities Referred by clinical teams or at the request of the Medical Director 20
Mortality Review South Tees Currently we are looking at 40% of all deaths in our trust 21
Mortality Review South Tees JAN 2017 Moved to secure, on line system Improved security, reliability and functionality and offers the opportunity to share our reviews with specialty teams more easily and potentially with other parts of the system including GPs and mental health trusts 22
North East secure, on-line mortality review system Basic Information pre-populates form PAS Demographics Date of admission Date of death Specialty Consultant Coding 23
North East secure, on-line mortality review system A range of review forms included for centralised review Locally adapted PRISM NMCRRP SJR Specialty Mental Health GP 24
North East secure, on-line mortality review system Grading of Care: 1. Preventability (PRISM/SJR) 2. NCEPOD Quality of care 3. PRISM Overall Quality scale 4. Expected or unexpected 25
North East secure, on-line mortality review system A qualitative narrative Automated reporting of quantitative and qualitative information 26
JAN 2016 A brief history in the North East 1 2 7 Trusts share aggregated data 7 12626 DEATHS 54% reviewed 0.4% preventable 3 6 5 4 27
A brief history in the North East MAR 2016 4 Trusts combine patient level review data Support of AHSN & NEQOS Single online database PAS data Case record review of process measures Grading of care Case narratives 28
A brief history in the North East 7370 REVIEWS 27 (0.4%) >50/50 chance of being preventable Single online database 20% judged to have room for improvement PAS in clinical, data organisational Case or both record aspects of care review of process measures Grading of care Case narratives 29
A brief history in the North East SEP 2016 Testing of on-line secure system commences 30
A brief history in the North East JAN 2017 Clarity Informatics Assure system is adopted in South Tees FEB 2017 IG discussions with Mental Health Trusts start in order to set up a pilot Second acute Trust adopts Assure system 31
Conclusions It is feasible to carry out large numbers of case record reviews and to link them locally to other governance processes including investigations of Serious Incidents. 32
Conclusions Case record review estimates of rates of avoidable death are subject to denominator effects. Internal reviewers, in our experience, report lower rates of preventable mortality than the published literature. 33
Conclusions Sharing information about deaths between providers means solving important IG issues. There are technical issues to overcome but the NHS Number could be used. Ideally we would use one system across a region. 34
Conclusions Medical Examiners are due to be introduced across England by April 2019. Some Trusts may implement interim arrangements to employ MEs as they offer major advantages including involving families in a timely manner. 35
PRISM 36
PRISM Results PRISM 1: Reviewers judged 5.2% (95% CI 3.8% to 6.6%) of deaths as having a 50% or greater chance of being preventable. PRISM 2: The proportion of avoidable deaths was 3.6% (95% confidence interval 3.0% to 4.3%). It was lower in 2012/13 (3.0%, 2.4% to 3.7%) than in 2009 (5.2%, 3.8% to 6.6%). There was a small but statistically non-significant association between HSMR and the proportion of avoidable deaths (regression coefficient 0.3, 95% confidence interval 0.2 to 0.7). The regression coefficient was similar for both time periods (0.1 and 0.3). This implies that a difference in HSMR of between 105 and 115 would be associated with an increase of only 0.3% (95% confidence interval 0.2% to 0.7%) in the proportion of avoidable deaths. A similar weak non-significant association was observed for SHMI (regression coefficient 0.3, 95% confidence interval 0.3 to 1.0). 37
North East Quality Observatory Service (NEQOS) Ridley House Henry Street Newcastle upon Tyne NE3 1DQ +44 (0)191 245 6708 neqos@nhs.net Follow us on @nhs_quality NEQOS is jointly hosted by Northumberland, Tyne & Wear and South Tees Hospitals NHS Foundation Trusts North East Quality Observatory Service (NEQOS) 2017 38