Plenary 3
Would I be comfortable in ICU?
Is it really safe?
Professor Cliff Hughes
A/Prof Tony Burrell
23 August 2014
Australia?
Australia?
Australia?
Home to seven of the worlds most deadly snakes!
The ICU is a dangerous place
High error ratesmost likely to occur in
intensive care units, operating rooms, and
emergency departments (Institute of
Medicine 1999)
Wide range of incidents, inappropriate
medical decisions, adverse drug events,
preventable slips & lapses (unrecognised
omissions in care) and variable
implementation of evidence-based care
The ICU is a dangerous place
High error ratesmost likely to occur in
intensive care units, operating rooms, and
emergency departments
Wide range of incidents, inappropriate
medical decisions, adverse drug events,
preventable slips & lapses (unrecognised
omissions in care) and variable
implementation of evidence-based care
ESPECIALLY FOR RUSSELL
Structure, process, outcome and
culture
Lilford Lancet 2004
Structure/Culture
Occupancy
After hours discharges = poorer outcomes
Goldfrad et al Lancet 2000
Tobin & Santamaria Med J Aust 2006
Pilcher et al Anaesth Intens care 2007
After hours admissions = ?poorer outcomes
Bhonagiri et al Med J Aust 2011
Medical staffing
After hours incidents
Nursing staffing
Increased nursing staff associated with less adverse
events
Kane et al Med Care 2008
Blegan et al Medical Care 2011
Process: How we do our business
Incident monitoring
Incident/problem recognised at the time
Medication errors
Incidents often not recognised & require different
approach
Checklists etc
Communications
Often fragmented, daily goals sheet
Emphasis on appropriateness of care -
housekeeping
Growing body of evidence linking improved
process of care with better outcomes
Outcomes
Morbidity & mortality meetings
Traditional
Often mandatory for training programs
Peer review - case by case discussion
The objectives of the surgical M&M conference are to
learn from complications, to modify surgical behaviour
and judgement based on previous experience, and to
prevent repetition of the problems leading to the
complication. (Murayama) i.e. emphasis on
teaching/learning
derive knowledge and insight without blame or
derision (quoted by Murayama)
Forum for discussion of major incidents
But not aggregated data
Risk-adjusted Outcomes
Crude hospital mortality rates are unsatisfactory
for measuring performance because do not adjust
for case mix or severity of disease
APACHE III, SAPS scoring systems adjust for
variation in patient outcomes that stem from
differences in patients and organisations ie
variations in casemix
Risk adjusted scores used to calculate
Standardised Mortality Ratios (SMR) - observed vs
predicted mortality allows benchmarking of ICUs
?helpful indentifying why ICUs are outliers
Outliers Data review
Data quality especially GCS
SMR of ventilated patients
SMR of acute pneumonia
SMR of cardiothoracic patients
SMR of deteriorating patients/medical
emergencies admitted from the wards
SMR end-of-life
Not much help
NSW Tertiary unit funnel plot (2009):
NB SMR
Outlier
Syrec study
79 ICUs in 76 hospitals in Spain
591/1017 (58%) patients suffered incidents (n=1224)
943 = no harm events
481 = adverse events temporary damage (29%) and
permanent damage or death (4%)
Causes:
Medication (74%)
Equipment (15%)
Nursing care (14%)
Accidental removal of vasc devices & catheters (10%) or
Airways & mechanical ventilation (10%)
Avoidable in 79%
Merino et al Int J Qual Health Care 2012
Other Work
Performance level failures most commonly preventable
slips and lapses
Rothschild et al Crit Care Med 2005
SEE Study 25.5% unintended events were ADEs
including wrong drug in 39/147
Valentin Intensive Care Medicine 2008
187 errors (3.3%) - vasoactive drugs (32.6%),
sedatives/analgesics (25.7%) & wrong infusion rate 71
(40.1%)
Calabrese et al Intensive Care Med 2001
Incidents by principal incident type - NSW ICUs 2010
NSW Medication errors q3 2011
Diazepam - 3
Frusemide - 3
Nimodipine - 3
Noradrenaline - 3
Pantoprazole sodium - 3
Amiodarone hydrochloride - 4
Cephazolin sodium - 4
Heparin - 4
Midazolam - 4
Oxycodone - 4
Vancomycin hydrochloride - 4
Heparin sodium - 5
Morphine sulfate - 5
Fentanyl - 9
Morphine - 11
Total 69
Primary problem as perceived
by the notifier
Administration - 79
Prescribing - 28
Narcotic controlled drug
discrepancy - 17
Dispensing - 12
Storage/Wastage/Security - 11
Delivery - 5
Supply/Ordering - 5
Undesired drug effect drug
therapeutic use - 2
Presentation - 0
Total - 159
Most errors go unreported
Many medication errors are not recognised many
are acts of omission
Only a fraction of ADEs are identified by incident
reports
One study examined 55 ADEs
15 preventable
26 serious or life threatening but only 2 had incident
reports
Conclusion voluntary reporting identified only a
small fraction of ADEs
Cullen et al Jt Comm J Qual Improv 1995
Competency
All staff are competent to provide care to the patients
at the level at which they have responsibility
procedures, decision making, supervision
Competency to assess technical and non-technical
skills
Airway management NB after hours
CVL insertion
Intercostal tube
Airway incidents 62.5% between 1600-0700
Why do medication errors occur?
High stress environment, fatigue, task saturation,
interruptions and reliance on MEMORY
Lack of knowledge
Medication safety management is complex
process prescribing, preparation - between 80
200 steps
Errors in order writing are the most common
medical error - illegibility, mistakes in
transcription
Patient identification ask Cliff
NSW CVL Incidents 2008-2011 n=572
Air Embolism
Incident report of death after removal of a
central line while patient was sitting up a
never event
Review of all incidents 2008-2014:
Reported cases
Two survived
6 died
RCA Causal statements:
Too difficult to put patient in bed
I knew about policy but I was too busy
Air Embolism
Air Embolism
Air Embolism
Checklists
Every day patients are not getting basic care
Routine care ie FASTHUG could happen
automatically and consistently with the use
of a checklist
Increasing evidence to show that checklists
are very useful in the ICU
Checklists: Daily goals
A checklist in the intensive care environment could have the
following advantages:
Immediate patient safety i.e. ensuring that the patient
gets what he/she needs immediately safety lesson from
aviation
Educational tool constant repetition reinforces the
principle e.g. BSL<8 a method of immediately
implementing evidence based practice
If the data collection process can be worked out e.g. PDA it
could be an effective method of collecting process
indicators. This approach requires the clinicians on the
round to address the issue and not a separate quality
process. It takes less than a minute to review the data.
Depending how developed improved communication
Pain
71% pts had pain assessed by medical team
on the study day
Of the 115 patients in pain, 42% did not have
pain score recorded <4 hrs
Pts who had surgery in past 4 days were more
likely than non-surgical pts to:
have pain assessed by medical team
(p=.001)
have a pain score documented in past 4
hours (p=.0001)
Pressure Areas
A pressure area risk assessment tool had not
been used in previous 24 hrs for 31%
110 pts (17%) had one or more pressure
areas, of these:
35% no risk assessment tool used
23% no targeted interventions
implemented
Elliott, McKinley, Fox Am J Crit Care 2008
Communication
Significant number of dangerous human errors
attributed to problems of communication
Donchin, Gopher, Olin et al Qual Saf Health Care 2003
Staff often unclear about management plan as
few as 10% residents & nurses in one study
daily Goals Sheet
Pronovost et al J Crit Care 2003
75% ward round time spent on communications
conversation-initiating interruptions occurred
at 14 /hour 37% communication time
Alvarez, Coiera Int J Med Inform 2005
Good Communication:
Accurate information exchange.
Enables us to:
Learn essential information
Share information
Form bonds
Foster understanding
Grow
Express our needs and feelings
Learn of others needs and feelings
Connect in meaningful ways
Communication In Health
Occurs between practitioners, patients, managers, consumers, community
Written emails, correspondence, newsletter, prescriptions, policies, posters,
noticeboards
Oral bedside manner, performance feedback, coffee room chat, managerial
style
Filtered implied vs intended
Often life-threatening consequences if get it wrong
In Health, multi-disciplinary
often means multi-lingual
Acronyms
Medi-speak
Nurse-speak
Allied-speak
Clinician-speak
Manager-speak
Patient-speak
Bureau-speak
Non-Verbals: c93% of the message
NSW Central Line Associated
Bacteraemia ICU Project
AR Burrell, M-L McLaws, A Pantle, M Murgo, E Calabria
Financial costs of CLAB
US estimates
15 680 lives and $1.3 billion medical costs could be saved annually by
reducing the number of CLABS*
United States House of Representatives Committee on Oversight and Government
Reform Staff Report September 2008, Survey of State Hospital Association:
Practices to prevent hospital associated bloodstream infections
The US Agency for Healthcare Research and Quality recently committed $3
million over 3 years to help reduce the incidence of CLAB
Health Care Advisory Board, Daily Briefing, 10 February 2008
NSW CLAB-ICU
Top down/bottom up project NSW Intensive Care Coordination & Monitoring
Unit and Clinical Excellence Commission
Methodology modelled on the work of Pronovost et al.
The project promoted a standardised insertion technique including:
Hand washing
Full barrier precautions during insertion
Cleaning skin with chlorhexidine
Avoiding femoral site if possible
Removing unnecessary catheters
Also included a retrospective review of all incidents entered into the NSW
Incident monitoring system
Guideline and checklist
8th April 2014
C F Hughes
Checklist detail
Minimum practical requirement
and assessment
Perform minimum of
3 supervised
insertions at each site
Perform
minimum of 5
independent
insertions
Observe
minimum of
1 insertion
Theory
assessment
Practical
assessment
Final
signoff
Continuingl
earning
Results
Data on 10,890 line insertions
Concurrent incident review:
Retained/lost guidewires
Arterial puncture
Multiple passes
Inadequately secured lines
Inadequate position check prior to use
Lack of access to ultrasound equipment
Policy breaches
Training & supervision common themes
Safety Alert for guidewires issued
Training framework developed
Checklist Compliance:
all ICUs July 07 Dec 08
Data on 10,890 line insertions
Competency assessed
48.3% (22.9% no, 28.8% missing)
Hat, mask, eyewear
79.9%
Hands washed 2 mins
91.6%
Sterile gown/gloves
95.9%
Alcoholic chlorhexidine prep allowed to
95.8%
dry
Entire patient draped
93.4%
Sterile technique maintained
95.6%
No multiple passes
80.9%
Confirm position radiologically
74.3%
Impact of compliance
Non compliance with the clinician bundle:
relative risk of CLAB was RR 1.62 (95% CI 1.1-2.4,
p=0.0178)
For central lines RR 1.99 (95% CI 1.2-3.2 ,
p=0.0037)
For PICC RR 5.08 (95% CI 1.03-25 , p=0.059)
Dialysis catheters no difference
If compliant with both clinician bundle and patient
bundle then risk of CLAB was RR 0.6 (95%CI 0.4-0.9,
p=0.0103)
Progress
Results
The single biggest problem in
communication is the illusion
that it has taken place.
George Bernard Shaw
Thank you
For further information:
[Link]
