Sepsis Management in Scotland Calum McGregor Consultant Acute Medicine National Clinical Lead for Acute Care Healthcare Improvement Scotland
Sepsis Management in Scotland Outline: Background on sepsis - Where did sepsis campaign come from? Why do we need it? Sepsis Collaborative 2012 to 2014 Post Collaborative Future Plans
Recognising Sepsis
Recognising Sepsis can be difficult Non-specifically unwell 87% of physicians symptoms of sepsis can easily be attributed to other conditions, causing late or mis-diagnosis (1) 81% - lack of a common definition could potentially lead to delays in treatment (1) No single test available
Sepsis Mortality by SIRS Rangel-Fausto et al. JAMA. 273(2):117-123, Jan 11, 1995.
Sepsis Standardisation Barcelona 2002 16 page document 46 recommendations Daniels R J. Antimicrob. Chemother. 2011;66:ii11-ii23
Care Bundle Approach IHI state: "a structured way of improving processes of care and patient outcomes. It is a set of practices that, when performed collectively, reliably and continuously, have been proven to improve patient outcomes.
Central Line Insertion U/S guidance Sterile gown and gloves Hat and mask Sterile field maintained Suture and dressing applied under sterile conditions
Central line infection rate since bundle implementation in 2008
SEPSIS 2012!
SEPSIS 2012 SEPSIS IN SCOTLAND SEPT 2012
Sepsis 2012 Waited 10 hours for a medical review Received 1gm Paracetamol and 500mls N Saline Disgusted and angry, completely let down by the system, felt my son s health was being compromised
Sepsis in 2012 Wishaw receiving unit Sept 2012: 0/11 received sepsis 6 Average time to antibiotic = 6 hours Nationally: 25% of patients with severe sepsis receiving IV antibiotics within an hour http://www.stag.scot.nhs.uk/ SEPSIS/Main.html
Rory Stauntan
If something good comes from Rory s death, it will be that we realize we have a broken system. Patient care is so fragmented. For the most part, medical professionals aren t taught the human skills that some deride as soft skills. So there s insufficient sharing of information and ineffective communication. Some in the medical field look upon these deaths as an unavoidable consequence of giving care. But they re inexcusable and unthinkable.
CHALLENGES Local and National A culture not receptive to quality improvement Improvement science not being applied consistently Unwanted variation Silo working Patient/carer voice not being heard
National Sepsis Improvement 2012-2014 HIS Breakthrough Collaborative AIM:10% mortality reduction Care Bundle Approach Sepsis 6 SAPG / HIS involvement Support front line teams
National HIS Driver Diagram for Deteriorating Patients
No of boards implemented National Support Structure NEWS implementation in Scotland's acute hospitals 14 12 10 8 6 4 2 0 2010 2011 2012 2013 2014 2015 2016 2017 Year
The aggregation of Marginal Gains
Reduce Unwanted Variation Antibiotics not in department Patient going to X-ray prior to antibiotics and fluids Triage system not robust enough to prioritise sick patients Nursing staff not informed of STAT antibiotic prescription MEWS added incorrectly Not applicable section on form Medical Students. Lack of awareness WE RE TOO BUSY
Reduce Unwanted Variation Brent James Intermountain Healthcare Identify a high-priority clinical process Build an evidence-based protocol (always imperfect: poor evidence, unreliable consensus) Demand that clinicians vary based on patient need Constantly update and improve the protocol
Small Tests of Change
DELIVER RESULTS Process Measures
Wishaw sepsis mortality ICD codes A40/A41
What s happened post Collaborative? UK Antarctic Research Vessel # NOF
What s Happened post collaborative? Sepsis Re-defined 2016 Sepsis: Life threatening organ dysfunction caused by a dysregulated host response to infection 2016 Third International Consensus Definitions for Sepsis and Septic Shock
Red Flag Sepsis
Q sofa Q sofa Criteria JAMA. 2016; 315(8): 762-774 SBP <100 R/R >22 GCS<15 2 or more assoc with increased mortality
Recommendations Scottish Patient Safety Programme - statement regarding the new international census definition of sepsis 1. The National Early Warning Score (NEWS) used for identifying patients, including those with sepsis. 2. NEWS trigger points for sepsis screening and management will continue to be locally defined. Screening for sepsis should be undertaken with the question could this deterioration be due to infection. 3. Systemic Inflammatory Response (SIRS) criteria will continue to aid in the general diagnosis of infection. 4. The qsofa criteria may be used as an adjunct to identify patients at increased risk of death and support decisions about treatment escalation. 5. All monitoring and screening tools should be viewed as an adjunct to clinical judgement. 6. Further studies on qsofa will inform decisions about their potential use as a screening tool for sepsis.
Prevention Harm Recognition Response System Enablers
Update -Spread
Primary Care / Scottish Ambulance Service Pre-Alerting in NHS Lanarkshire, GG and C, Highland and Grampian Reliable NEWS completion and communication with secondary care Martin Carberry, and John Harden BMJ Qual Improv Report 2016;5:u212670.w5049
PRE-ALERTING Improve Patient Journey
Mr C timeline Call received 1223 Call passed 1243 Crew at scene 1311 Arrival at WDG 1332 SEPSIS 6: Time zero - 1332 Completed - 1344
Time to Treatment and Mortality during Mandatory Care for Sepsis (Rory s Regulations) - June 2017 49000 patients with severe sepsis / septic shock New York State 4% Increased mortality for every hour delay in antibiotic (odds ratio: 1.04 per hour; 95% CI, 1.03 to 1.06; P<0.001) 14% mortality increase after 3 hours NEJM - 2017;376:2235-44
Small changes can make a big difference!
Future Plans Keep reducing harm from sepsis Build on the success of past 5 years Continue MDT working across traditional boundaries Pre-alerting Spread to other areas
Summary Reduced mortality from a national quality improvement programme Effective care bundle Accurate identification of sepsis remains a challenge Breaking down traditional barriers key to improvement work
Any Questions?