Transcript Diagnostic Decisionmaking – META Scholars

Diagnostic Decision-Making:
How do we do it and how can we
(and our learners) improve?
META Scholars
September 5, 2013
Agenda
• Overview of diagnostic reasoning
• How good are we?
• How can we (and our learners)
improve?
Objectives
• Be able to describe the basic process of
making a diagnosis
• Acknowledge we struggle with making
diagnoses
• List several ways we can improve our
diagnostic skills
Overview of Clinical Reasoning
• Overview of making a diagnosis
• How our brains deal with it
• What it actually looks like in practice
How do Doctors Think?
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Data collection
•
•
•
•
History
Physical examination
Laboratory studies
Imaging studies
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Problem Representation
•
•
•
•
•
Making sense of the data obtained
Identification of the key elements
Categorization
Semantic qualifiers
Frame things (context is everything)
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Illness Scripts
• Mental representations of the key
elements of specific diagnoses
– History
– Physical
– Labs
– Imaging
– Response to therapy
Acute Coronary
Syndrome
Pericarditis
Pulmonary embolism
Aortic Dissection (AD)
Epidemiology
Older age, risk factors
include diabetes,
hypertension,
dyslipidemia, family
history, tobacco use
Uremia, auto-immune
disease, prior URI,
recent MI or heart
surgery, malignancy
Risk factors of
endothelial injury,
hypercoaguability, and
stasis: recent surgery,
active cancer (e.g.
adenocarcinoma),
medications (e.g. OCP);
immobility
Older patient, HTN the
primary risk. Younger
patients also at risk
(cocaine, collagen
vascular, bicuspic aortic
valve…)
Time Course
Acute onset, not
necessarily preceded by
exertional angina
Acute, but may occur
in setting of sub-acute
or chronic disease
Acute onset usually
without progression,
unless second PE
Acute onset, usually
constant
Clinical Features
(1) History
(2) Exam
(3) Labs
(4)Imaging
Advanced Studies
1) Chest pain, with
crescendo to maximal
pain; often dull and substernal, radiating to
arms/shoulders;
diaphoresis; dyspnea;
nausea/vomiting,
diaphoresis.
2) Tachycardia
3) Elevated cardiac
biomarkers (troponin/CK),
abnormal ECG (ST
elevation/ depression, T
wave changes)
4) Regional wall motion
abnormality on
echocardiogram
1) Sharp, stabbing
chest pain radiating to
back and trapezius
ridge; improved with
sitting forward
2) Pericardial friction
rub (may be
ephemeral, more
pronounced with sitting
forward)
3) Abnormal ECG
(diffuse ST elevation,
PR depression);
elevated inflammatory
markers (ESR, CRP)
4) Common: Pericardial
effusion on echo or CT
1) Shortness of breath,
pleuritic chest pain
2) Tachycardia;
tachypnea; normal lung
exam,
3. Common: positive Ddimer
4. Xray with minimal
abnormalities; CT chest
with pulmonary
angiogram demonstrates
a clot; V/Q scan with
unmatched perfusion
defect
1) Common: Sudden
onset, severe ripping and
tearing CP radiating to
back
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Illness Script Selection
• Match the problem formulation to the
illness script
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Overview of Clinical Reasoning
• Overview of making a diagnosis
• How our brains deal with it
• What it actually looks like in practice
How do doctors think?
• We’re not really sure, but we do have a
general idea
• A couple of key points:
– Experience really matters
– Lots of complexity
Question 1:
Image from Wikimedia Commons
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Question 2:
Data Collection
Problem
Representation
(Framing)
Potential
Match
Diagnosis!
Access Illness Scripts
Overview of Clinical Reasoning
• Overview of making a diagnosis
• How our brains deal with it
• What it actually looks like in practice
How it plays out….
• Bedside Clinical Reasoning
– Hypothesis generation
– Hypotheses refinement
– Diagnostic testing
– Causal reasoning
– Diagnostic verification
A Case
• 69 year-old man with a history of CAD
presents with chest pain
– Acute coronary syndrome!
Hypothesis Generation
• Unlike prior MI
• Pain is sharp and stabbing
– Less likely ACS, maybe PE?
– Pericarditis?
• No associated dyspnea
• Radiates through to the back
– ?Aortic Dissection
Hypothesis Refinement
and Generation
• Exam
– Differential pulses in
upper extremities
– Aortic insufficiency murmur
Causal Reasoning
Hypothesis
Refinement
• CXR
– Widened mediastinum
• CT scan
– Aortic dissection
Diagnostic Testing
and Verification
• Bedside Clinical Reasoning
– Hypothesis generation
– Hypotheses refinement
– Diagnostic testing
– Causal reasoning
– Diagnostic verification
Agenda
• Overview of diagnostic reasoning
• How good are we?
• How can we (and our learners)
improve?
Definition of a Diagnostic Error:
• A diagnosis that, on the basis of the
eventual appreciation of more definitive
information, was
– Unintentionally delayed, or
– Wrong, or
– Missed altogether
Question 3
What is your personal rate of diagnostic error?
A)
B)
C)
D)
E)
<1%
2-3%
5%
10-15%
>20%
Question 4
What is the overall rate of diagnostic error in
medicine?
A)
B)
C)
D)
E)
<1%
2-3%
5%
10-15%
>20%
Rate of Diagnostic Error
• Overall, likely rate of diagnostic error is
about 10%
• Error rate varies by specialty and study
– Anatomic pathology 2-5%
– ED up to 12%
– Medical inpatient diagnosis ~6-8%
Do these errors matter?
• Account for up to 17% of adverse events
• 40,000-80,000 US hospital deaths per
year attributable to diagnostic error
• 5% of all autopsies show a lethal
diagnosis that could have been treated
ante-mortem
• Tort claims data (really expensive)
JAMA 2002; 288:2405
What do these errors look like?
Diagnosis
Stroke
Sub-arachnoid
hemorrhage
Pulmonary Tb
Missed on initial
evaluation
9%
5%
45%
Acute Coronary
Syndrome
2-3%
Appendicitis
19%
What causes these errors?
• Three general categories of diagnostic
error
– “No Fault” (7%)
• Very unusual presentations, patient-related
error
– Systems-related (19%)
46%
• Technical failure, organizational issues
– Cognitive errors (28%)
• Faults in knowledge, data gathering, information
processing or metacognition
Arch Intern Med 2005;165:1493-1499.
Basis of Cognitive Errors
• Cognitive Errors
– Faulty knowledge
– Faulty data gathering
– Faulty synthesis
– Affective error
Basis of Cognitive Errors
• Cognitive Errors
– Faulty knowledge
– Faulty data gathering
• Failure to ask or look
• EMRs
– Faulty synthesis
– Affective error
Red Flag Medicine
• We often embrace “Red Flag Medicine”
– Overly trusting of technology
– Doubt the utility of the clinical exam
– Lack confidence in clinical skills
!
Basis of Cognitive Errors
• Cognitive Errors
– Faulty knowledge
– Faulty data gathering
• Failure to ask or look
• EMRs
– Faulty synthesis
– Affective error
Basis of Cognitive Errors
• Cognitive Errors
– Faulty knowledge
– Faulty data gathering
• Failure to ask or look
• EMRs
– Faulty synthesis/metacognition
• Premature closure
• Misjudging the importance of a finding
• Faulty context generation
Question 5:
• List two things that annoy you about
people
• List three of your favorite people
Basis of Cognitive Errors
• Cognitive Errors
– Faulty knowledge
– Faulty data gathering
– Faulty synthesis
– Affective error
Agenda
• Overview of diagnostic reasoning
• How good are we?
• How can we (and our learners)
improve?
Potential Solutions
• Monitoring and feedback systems
• Reframe root cause analysis
• Provide improved clinical decision
support
• Mandate/encourage appropriate use of
EMRs
• Data visualization tools
• Cognitive awareness and techniques
Expert
Performance
Experienced
Non Expert
Time
Slide from Gurpreet Dhaliwal
Making Experts
•
•
•
•
Progressive Problem Solving
Feedback
Simulation
Deliberate Practice
Progressive Problem Solving
• Avoid the routinization of work
– Go past where you have to
• Reformulate problems
– Add challenging, nuance and difficulty
Diagnostic Feedback
• Diagnostic Closure
• Are we really as good as we think we
are?
Croskerry P. The feedback sanction. Academic Emergency Med 2000.
Simulation
• Practice, practice, practice
• We can’t see as many patients as we
need to
• We don’t see all the presentations and
diseases we need to
High-Fidelity Sim
Fox MC et al. N Engl J Med 2013;369:966-972
Deliberate Practice
•
•
•
•
What do I stink at?
Focus on it
Work on it repeatedly
Assess performance
Fox MC et al. N Engl J Med 2013;369:966-972
Habits for Good to Great
Experienced
Expert
As needed
Progressive Problem
Solving
Feedback on my patient outcomes
Random
Sought out
Case Reading
Spectator
Simulator
As it happens
Deliberate Practice
On The Job Learning
Skill Development
Dhaliwal G. Clinical Excellence: Make It A Habit. Academic Medicine 2012
Action Steps
1. Mindset
 Continuous
learning/pushing
ourselves
2. Feedback
 Set up a system
3. Simulation
 One case per week
4. What is lacking?
 Get deliberate
Slide from Gurpreet Dhaliwal
Question 6:
• List the two most important things you
learned in the past hour
• List the two things you wish we had
covered but didn’t
Agenda
• Overview of diagnostic reasoning
• How good are we?
• How can we (and our learners)
improve?
Objectives
• Be able to describe the basic process of
making a diagnosis
• Acknowledge we struggle with making
diagnoses
• List several ways we can improve our
diagnostic skills
More Information
http://www.improvediagnosis.org/?Clinical
Reasoning
Diagnostic Decision-Making:
How do we do it and how can we
(and our learners) improve?
META Scholars
September 5, 2013