1. Eliminating Harm:
The Foundation of High Performance
in Healthcare
Richard P. Shannon
Frank Wister Thomas Professor of
Medicine
Chairman, Department of Medicine
University of Pennsylvania
School of Medicine

2. A Metaphor for the US Healthcare System
This is a transcript of an actual radio conversation between a US
Naval ship and Canadian authorities off the coast of
Newfoundland in October, 1995. Radio conversation released by
the Chief of Naval Operations 10/10/95
• Canadians: Please divert your course 15
degrees to the South to avoid collision.
• United States: Recommend you divert
your course 15 degrees to the North to
avoid a collision.

3. • Canadians: Negative. We insist that you
divert your course 15 degrees to the South
to avoid a collision.
• United States: This is the Captain of a US
Navy Ship. I say again, divert YOUR
course.
• Canadians: No. I say again, you divert
YOUR course.

4. • United States: This is the aircraft carrier USS
Lincoln, the second largest ship in the United
States’ Atlantic Fleet. We are accompanied by
three destroyers, three cruisers, and numerous
support vessels. I demand that you change your
course 15 degrees North! I say again, that’s one
five degrees North or counter-measures will be
undertaken to ensure the safety of this ship.

5. • Canadians: This is a
lighthouse. Your call.
Photo by Lynn Botterman

6. Our Contract With Society
• Commitment to professional competence
• Commitment to honesty with patients
• Commitment to patient confidentiality
• Maintenance of appropriate relationships with patients
• Commitment to improving quality of care
• Commitment to improving access to care
• Commitment to scientific knowledge
• Commitment to trust by managing conflict of interest
• Commitment to professional responsibilities
• Commitment to the just distribution of finite resources
ABIM Physician Charter

7. Primum Non Nocere

8. Oath of Hippocrates
Will prescribe regimens for the
good of my patients according to
my ability and my judgment and
never do harm to anyone
I will apply dietic measures for the
benefit of the sick according to my
ability and judgment; I will keep
them from harm and injustice.

9. The Question
• Can the elimination of harm (hospital
acquired infections, medication errors,
readmissions) serve as a starting point
for reducing unnecessary costs (waste) in
healthcare?
• Does it fulfill our professional duty to “do
no harm” and to be good stewards of
finite resources?

10. Healthcare Spending and Social Good
• US spends 18% of the GDP in healthcare
• CMS accounts for 20% of the total government
spending
-8x more than on education
-12x more than food aid
-30x more than on law enforcement
-78x more than conservation
-87x times more than water supply
-830x more than on energy conservation

11. Problems with the US Healthcare
System
• Costs are too high and value is too low
• Cut your way or improve your way to greater value
• If one can identify and eliminate waste, we can spare
cuts to important services
• Waste
overtreatment
failures in care delivery
failures in care transitions
excess administrative costs
fraud and abuse

12. High Performing Organizations
• High performing organizations are the
best in class
• They achieve high performance not
necessarily through technological
advances but through complete
engagement of all the wisdom and skill
embedded in each worker
• These organizations and their leaders
never stop learning
Spear S Chasing the Rabbit
Spear Chasing the Rabbit

13. Dynamics of HPO
• Cope with complexity by continuous
focus on learning more about how to
improve the work they do.
• Nothing is ever good enough
Spear S Chasing the Rabbit

14. The Four Capabilities of HPO
• Specifying work to capture existing
knowledge
• Swarm and solve problems to build new
knowledge (avoid “information
perishability”)
• Share that knowledge throughout the
organization
• Lead by developing these capabilities in all
workers Spear S Chasing the Rabbit
Spear Chasing the Rabbit

15. Leaders in HPOs
• Set clear and unambiguous expectations
• Practice not espouse values
• Amazing problem solving capabilities
• Empower and create systems that
“discover” the right answers
• They do away with excuses as to “why
not” or “if only, then”
Spear S Chasing the Rabbit

16. US Navy’s Nuclear Submarine Program
• 200 nuclear powered ships launched
• 5,700 reactor years of operation
• 154 million miles underway
• Not a single reactor related casualty or escape
of radiation
Spear S Chasing the Rabbit

17. Current US Estimates
• 5-10% of inpatients acquire an HAI
• 1.7 million HAIs annually
• 99,000 deaths
• Estimated costs:$28.4-33.8 billion
• It is 27X safer to work at Alcoa than it is to
walk into a US hospital
Safer?

18. HAI in Pennsylvania 2009
23,287 HAI (1.2%)

19. Patient Outcomes

20. Seven Leverage Points:
If you want to achieve system-level results…
1. Set specific system-level aims and oversee their
achievement at the highest levels of governance
2. Build an executable strategy to achieve the aims,
and oversee the execution at the highest levels of
administration
3. Channel attention to system-level aims and
measures
4. Get patients and families on your team!
5. Engage the CFO in achieving the aims
6. Engage doctors in achieving the aims
7. Build the improvement capability necessary to
achieve the aims Reinertsen IHI

21. A “Project”

22. System Level Aim

23. Setting the Course
Current Conditions
Root Cause Analysis
Decode: 37 CLABS
(July 2002-June 2003) Solve to root cause in real time
the origins of CLABS in
PRHI Central Line Data MICU / CCU
Observations of Dressing
Changes
Eliminate
CLABS
In MICU/CCU Counter Measures Generated
In 90 days By the People That Do The Work
Generate Additional
Reassess Results
Counter Measures

24. Problems With Bench Marking
The Difference Between Reporting and Actionable Data
10
9
8
7
6 CDC
5 CCU/MICU
4
PRHI
3
2
1
0
01 Q3 01 Q4 02 Q1 02 Q2 02 Q3 02 Q4 03 Q1 03 Q2

25. Decoding the Data:
What Does 5.1 infections/ 1000 line days
Really Mean??
• 37 patients / total of 49 infections
• 193 lines were employed (5.2 lines / patient)
• 1753 admissions
• 1063 patients had central access for more than 12 hours
• 1 out of 22 patients with a central line became infected.
• We were reporting only half the actual infections (not including
femoral line infections!!)
• Two-thirds of the infections involved virulent organisms. Twenty
percent were MRSA
• 19 patients died (51%)
Journal of Quality and Patient Safety 2006;32:479

26. Personal Stories Send the Message
• 22 yo. woman, a single mother of a 2 year old child, presented with
relapsing acute myeloid leukemia.
• Following re-induction with a highly toxic chemotherapy regimen,
she is found to be in complete remission.
• Day 18, she develops fever, chills and hypotension. BC grow staph
aureus from her Hickman catheter.
• In retrospect, the unused lumen of her triple lumen catheter had
cracked and been repaired.
• The cracked lumen-repair process was common place despite
evidence that it was associated with a27% risk of infection
• RCA revealed unspecified understanding about flushing unused
catheters and that there was a small area on the lumen where a
clamp should be re-enforced.
• The patient spend an additional 17 days in the hospital, away from
her child.
• She died 27 days after discharge.

27. Penn Medicine/ DOM
Approach to Patient Safety
• Trained 220 nurses in the Lean methodology
• Exposed all senior leadership to observation
exercises at the point of care
• Completely redesigned standard methods for
placing and maintaining all catheters.
• Incorporated training modules for house staff
and fellows.
• Create UBCL teams with Problem Solving
Skills

28. Disciplined Problem Solving
Background: What’s the problem or concern?
Current Condition Target Condition
How work is currently done and what problems prediction of the expected effects of
are encountered countermeasures
Root cause analysis Countermeasures
Factors revealed by investigation to cause or Changes in how the work is done to offset
contribute to the problem causal effects
Actual Outcome Expected Outcome
how the system performs with changes made
How did you expect the system to perform
Gap Analysis
difference between what was
28 predicted and what actually happened

29. Rounding on Sick Systems
Rounding on Sick patients Rounding on Sick Systems
• Chief complaint • What’s the problem ?
• Present illness • How is work currently
• Physical exam/diagnostic done?
test • What defects are
encountered in the work?
• Intervene to eliminate
• Therapeutic intervention
defects
• Clinical course
• Create a target condition
• Natural history
• Measure what actually
happens
• Assessment of outcome • Gap analysis
29

30. Variation in the Course of Work
(Line Placement)
• No standard pre-procedure checklist
• Informed consent in 25% of procedures
• Eight different ways to “gown and glove”
• Six different ways to “prep and drape”
• Four different approaches to central veins
• Five different insertion kits
• 55% of procedures were documented

31. The Current Condition of Variation
Steps 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
Enter Glove V isual Set up Wipe Flush Draw Wipe V ac Draw Draw Prob. V ac Wipe Flush Prob. Wipe Cap DiscardGlove Flush Wipe V ac Draw Wipe Draw Draw V ac Wipe Flush Wipe Cap DiscardLabel Docum Seal
. Wash
Glove Gow n Enter A larm Set up Cap Wipe V ac Draw Draw Draw A larm V ac Wipe Flush Wipe A larm Glove Label Docum send
.
Enter Set up Wipe Flush Waste Draw Wipe Draw Trans. Draw Draw Draw label Seal Docum Seal
. Glove Exit
Labels Supplies Enter Set up Purell A larm Disc. Tub
alcohol Un- CapBD Wipe V ac Draw Draw Draw Draw V ac Wipe Flush Discardlabel Seal Prob. Supplies larm
A alcohol Disc. Tub
Un- CapWipe V ac Draw Draw Draw V ac Flush DiscardDisc f lush
Wipe Rec. Tub
label Seal DiscardExit Wash send
Supplies Labels Glove Enter Med Set up Wipe Flush A larm V ac Draw Draw Draw V ac Wipe Flush Prob. (untag- line)
DiscardGlove A larm Label Docum Seal
. send
Supplies Enter Glove Unw rap larm
A Flush V ac Draw Draw Draw V ac Wipe Flush Rec. Tub iscardlabel
D Wash send
Enter Glove V isual Set up Wipe Flush Wipe V ac Draw Draw Draw Wipe V ac Wipe Flush Cap DiscardGlove Label Seal Wash send
Enter Glove V isual A larm Wipe Flush Prob.( no blood) ac
Flush V Draw Draw Draw V ac Wipe label Discard
Supplies Enter Glove V isual Set up(bed)
A larm Flush V ac Draw Draw Draw Draw V ac Wipe Flush Wipe Flush Glove Docum Label
. Seal
Labels Supplies Purell Glove Enter A larm Set up(bed) Tub
Disc. Unw rapWipe Flush Draw Wipe DiscardDraw Trans. Fill Fill DiscardWipe Flush Rec. Tub iscardlabel
D Glove Seal send Wash
Labels Supplies Glove Enter A larm Set up clamp Disc. Tub
Wipe Flush Draw Wipe V ac Fill Fill Fill Fill clamp V ac Wipe Flush clamp Rec. Tublam
c p DiscardA larm label Exit send Glove
Labels Supplies Label Docum Gow n
. Glove Enter Set up (TOWE Wipe
A larm L ON BE Flush
D) Draw Wipe Draw Needle Wipe Draw Needle Wipe Flush DiscardSyr Fill Fill Syr Fill Fill Fill DiscardGlove Exit Wash Seal send
Labels Supplies Unw rapDocum Wash
. Enter Set up(bed)
Glove Wipe Flush V ac Fill Fill Fill Fill Fill Fill V ac Wipe Flush Discardlabel BD label BD Wash Exit Docum Seal
. send
Labels Supplies Enter Unw rapSet up(bed)
Wipe Flush DiscardWipe Draw Trans. Fill Fill DiscardWipe Draw Trans. Fill Fill Fill DiscardWipe clamp Flush DiscardPurell Exit Docum Label
. Seal send
Labels Supplies Wash Unw rapGow n Glove Enter A larm Wipe Flush Draw Draw Wipe A larm DiscardSyr Fill Fill Fill Syr Fill Fill label BD DiscardGow n Glove Exit Seal send
Enter Gow n Purell Glove Supplies rob.
P Glove (of f ) n (of f )
Gow Wash Exit Supplies nter
E Set up(podium Gow n
Purell ) Glove Stock Explain A larm Un- CapStopcockac
V Fill Fill Flush Cap Stopcocklarm
A Gow n Glove Wash Exit Glove Label Seal send
Gow n Enter Glove Supplies larm
A Set up(bed)
Stopcock
Un- Cap ac
V Fill Fill Fill Fill Fill Syr Flush wV ac
ith green tube) Discardlabel
Cap Gow n Glove Purell Exit Seal send
Supplies Un- Cap Wipe Syr Fill Fill Wipe pump DiscardSyr recap
Order Label
Purell Glove Supplies xplain A larm
E Line Set up(bed) Cap
Un- (A BG &4x4)
Draw wDraw AWaste (gauze) Cap
aste BG Flush Flush DiscardPurell Spec/glove
Labels Label Seal send
Purell Labels Supplies
Glove A larm Line Set up(bed) 4x4)
A ssem Un- ac Cap/gauze
bleV Cap V ac Draw wFill (purple)
aste Flush wDiscardlabel
ith green tube) DiscardWash Spec/glove
Prob. Tries to Seal
send send
Glove Supplies Set up(bed) Cap ac
Un- 4x4)V Draw wFill red tube
aste A larm A BG syr
Cap A BG
Flush DiscardV ac Glove Spec/glove
Purell Labels Label Seal send Seal send
Comp. Purell Supplies ssem A V ac
A bl larm Glove Stopcock
Un- CapCap/ inf V ac
usion pum
Draw wFill (tiger top)
p aste V ac Flush DiscardCap stopcock
stop in A larm
vac DiscardGlove Purell Labels Label Seal send
Set up Flush V ac Fill Fill Flush Un-Gow n
Supplies
Wipe Draw Fill Flush Wipe Syr Fill Discard
Enter Wash Supplies
Wipe Unw rapDraw Trans. Wipe bottle
Fill Wipe bottle 2
Fill Wipe V ac Fill Wipe Unw rap f lush
Wipe Flush Wipe Syr Wipe Rec. Tub larm
A DiscardLabels Label Seal send
Supplies Gow n Glove Wipe Flush Wipe Supplies
Gow n Glove Wipe V ac Fill Syr Wipe Flush Discardbed TPA
Labels Supplies Gow n Glove Set up E
P rob.(l xplain Supplies larm Line
A StopcockUn- CapAssemble v Fill
V ac ac Fill Fill
Discar Stopcockvac Gauze Squeeze
on port Cap
Discar Discar Stopcock
Discar Flush Discardwaste A larm Discardlabel
gauze Discardother work
UN-Glove
Wash Glove Seal send
ook Suppli Set up Unw r Waste d d d d De-
Glove Gow n Enter f or
P rob.(l A larm Glove es (bed) ap
Waste V isual V ac (tube) Fill Fill
A larm Fill
Discar (vacut Syr
Discar Discar Flush Cap (sharp (red) (w hite UN-Glove
Gow n Wash Label Seal send
ook (10 (resta d d d De-
Glove Gow n Enter f or Flush V isual A larm Flush cc.) Draw Stopcock
Cap
Syr Flush
Prob. rt)
Syr. (sharp (red)
Prob. (w hite UN-Glove
Gow n
Cap Wash Label
Discar Seal
Discar send
Suppli Set up Stopc Cap (remo (cap (in (dirty (back d d De-
Order Gow n Glove Enter es (bed) V isual ock Syr Waste (hold)
V ac ve) on stopc Draw cap Unw rap 4x4
Flush on)
Discar Discar Discar Label Flush
Label (sharp (red) Gow n
Prob. Purell
call Label Seal send 1st 2nd
Set up Unw r Rem o (rem o d d d (pink, (pt., at (no blood call tube tube
Wash
P rob. Supplies Glove (bed) ap up
Set ve air
assem Wipe V ac
Cap Fill Fill ve) Wipe Flush Clamp A larm (sharp (red) (w hite supply desk) Docum Label
. tube) bank wait CDR wait arrive Seal send arrive
(look f or Suppli Stopc room ble Stopc (rem o Stopc Stopc
labels) Glove es ock
Prob. (Move vac
Suppli ock ve) V ac Fill Fill ock Flush
Prob. Cap
Prob. Flush ock Unw r Clear Discar Discar
Suppli (look es Set up Syr (3 (dif f . (m anip Syr ap air d d UN-
Purell Glove es
Prob. f or UN-Glove
(Omni Purell Glove A larm (bed)
Unw r Cap cc) draw ulate (A BG) Draw 4x4 Flush
Clear Waste Cap
Discar Discar f rom (w hite (sharp Label Glove Purell
(w ait Set up ap Syr (3 Syr Cap 4x4 air d d UN-
Order Label
Purell f or Purell
Unw r Glove
Unw r A larm BD (bed) Cap 4x4 cc)
Prob. Waste (A BG) (hold)
Cleare Cleare Discar (hold)
Discar Waste Cap f rom (sharp (red) Glove Purell Label Seal send
Suppli ap ap (m anip d line d line d d UN-
Purell Glove es 4x4 4x4 Cap Waste Fill Fill
V ac Fill ulate at at (sharp (red) Glove Purell Label Seal send
Set up (rem o
Supplies (table) Wipe V ac Waste Fill Fill Fill ve) Wipe Flush Clamp label BD Seal
Fill send
Suppli Docum . Discar Discar Discar
Unw r Rem o UN- (w ast es Set up d d d
Enter Explain A larm Wash Glove
Set up room (Map iv pole air ove chair)
Supplies (w ipes in - m Wipe
Fill ove
Discar ve room sse
A) UN-CapFlush Clamp V ac e) (stude (bed) Fill Fill V ac Flush ClampWipe (sharp (red) (w hite UN-Glove
Wash other work
Label Seal send
(w ast d m ble
Purell Wipe Clamp A ttach Flush
unclam Flush
p e)
Fill (sharp
Discar E xplain V ac Fill
Discar Fill Fill BD Wipe Flush Clamp Docum Wipe
. Disconnect tubing
Explain Label Seal Prob- no tubes told sec.
send
(w ast d d
Supplies Wash Glove Set up room (MFlush pole - m (sharp clam
Wipe ove iv e) ove chair) p Fill (sharp Flush Reconnect tubing
Docum Cap
. Discar Explain label Seal Prob- no tubes sec calling f or tubes
d
Supplies Purell Enter Glove (P larm Flush
Aurple) Wipe vac Waste Fill A larm Fill Flush A larm (sharp UnGlove urell
P labels Label ice Seal send
Discar
d
Wash Enter Set up bed
Glove prob Supplies
Wipe vac Flush Waste Fill Fill BD Fill label Wipe Flush clamp Wipe Flush
Discar label
Discar (sharp E
Discar xit Wash
d d d
Labels Supplies Glove Gow n Set up room table
A sem Unw rap
ble vac Wipe Flush Waste Wipe vac Fill Fill Fill Unw rapWipe Flush clamp (sharp (red) (w hite BD label

32. DRESSING CHANGE STANDARD WORK
1) Set up work Hand Hygiene Open Drape Open Dressing Kit Drop Biopatch
work Wash or Purell
Space
2) Prepare Adjust Bed Don Masks Clean Gloves
People (nurse)
(patient)
3) Remove Remove with alcohol Discard Trash Wash Hands
Dressing
4. Clean site Apply Chloraprep Allow to dry
Sterile 30 seconds 30 seconds
gloves
5. Apply New Apply Outline Apply Seal Apply Strips in
Biopatch Dressing Dressing Dressing X and label

33. Number of Infections Central Line Associated
Blood Stream Infections
20
15
10
5
0 l
t
t
ly
n
ri
p
p
a
u
p
e
e
J
J
S
S
A
Months

34. Approaching the Theoretical Limit?
The Journey of 1,000 Days
BSI
400
300
Infections
200
100
0
2007 2008 2009 2010

35. Problem Solving: The 5 Whys
• What did the patients get infected?
Defective sterilizing process
• Why was the process defective?
It was a rapid process that avoided ETOH rinsing
and medical air drying
• Why were we using the device?
It had the fastest turn around time
• Why did we need a fast turn around?
We were doing a lot of bronchoscopies
• Why were we doing a lot of bronchoscopies?
We had many ventilator associated pneumonia

36. Eliminating VAP:
How Did We Do It?
• Step 1: Elevate the head of the Bed 30°
• Step 2: Chlorhexidine mouthwash BID
• Step 3: Change vent tubing weekly
• Step 4: Change suction catheter daily
• Step 5: provide a hook for hanging resuscitation
bag
• Step 6: Check endotracheal cuff pressure
Total Added Cost: $17/ ventilated patient

37. Applying a Construct:
Placing, Maintaining, and Manipulating
50
E-VAP
40
FY 04
FY 05
30
FY 06
FY07
20
FY08
FY09
10 FY10
0
FY 04 FY 05 FY 06 FY07 FY08 FY09 FY10

38. Reductions in HAIs
60
50 BSI
VAPS
40 MRSA
30
20
10
0
2001 2002 2003 2004 2005 2006 2007 2008
Fiscal Year
Journal of Quality and Patient Safety 2006;32:479

39. Improvement is Based on Values
• Improvement is about Values not about
tools and technology
• Three Core Values of Effective Improvement Teams
3. I am treated with dignity and respect by all
regardless of my education or rank
4. I am given the tools to make a contribution that adds
meaning to my life
5. I am recognized for what I do.
• Practice not espouse Values

40. Values Trumps Process

41. The Conspiracy of Error and
Waste
• What is the cost of a CA-BSI in human
and financial terms?
• What does society pay for healthcare
associated infections (HAI)?
• Do hospitals and physicians make money
on HAIs ?

42. Case 1:
• 37 year old video game programmer, father of 4,
admitted with acute pancreatitis secondary to
hypertriglyceridemia.
• Day 3: developed hypotension, and respiratory failure
• Day 6 : fever and blood cultures positive for MRSA
secondary to a femoral vein catheter in place for 4
days.
• Multiple infectious complications requiring exploratory
laparotomy and eventually tracheostomy
• Day 86: Discharged to nursing home

43. The Losses Attributable to CA-
BSI are Staggering
• Average Payments: $64,894
• Average Expense: $91,733
• Average Loss from Operations: -$26,839
• Total Loss from Operations:-$1,449,306
• In only 4 cases did the hospital make money!
• The cost of the additional care averaged 43% of the
total costs of care
• Average LOS: 28 days (7-137)
• Only three patients were discharged to home.

44. The Losses Attributable to Ventilator
associated Pneumonia are Equally
Staggering
• Average Payments: $62,883
• Average Expense: $87,318
• Average Loss from Operations: -$24,435
• Total Loss from Operations:-$2,419,065
• The average payments were twice that for a similar care
without VAP ($33,569)
• Average LOS: 34 days versus 17 days
• 32% of patients died and 43% underwent tracheotomy.

45. CCU/MICU and HAI
A Big Return on Investment
• Total Operating Improvements
CLAB= $1,235,765 (2 years)
VAP= $1,003,162 (1 year)
MRSA= $ 295,342 (1 year)
• Highmark PFP = $3,100,000 (2 years)
• HAI elimination Initiatives = +$5,634,269
• Investment = $85,607
• 388 additional ICU admissions
• 57 lives saved

46. Strategies for Reducing Per Unit Cost
(Pugh)
Traditional Quality *waste = unintended variation,
Strategy: rework, error, valueless care,
Strategy:
Control Inputs needless complexity, etc.
Remove Waste*
Direct Inputs from Production
•Supplies Big Dots
Clinical •Financial
•Labor •Clinical
Processes
•Patient Experience
Indirect Inputs
•Structure Support
•Technology Processes

47. Build a Parking Garage or Fix the
Care Process?
Not more…better Not volume….value

48. Modifying the Patient Experience
4:05 $32 parking
77 min
22 31 23 18 17 14 18
14 days 57 min
min min min min min min 97 min min
Call Wait for App Travel Park Reg Wait VS Wait MD CO Tests Exit
15 7 20 11 14
7 days 57 min min min min
45 min
min min
37 min
2:02 $8 parking
↑Patient visits from 7-9/session
↑ On –Time Performance
↑ Patient satisfaction (waiting)
↓ Lag days

49. Summary
• Data must not only be reportable but actionable
• Lessons borrowed from HPO are widely applicable in
health and medical care
• HAI are not inevitable, but rather fully preventable
• The elimination of unsafe conditions such as HAI will
free up extensive financial resources currently
consumed in their care.
• In eliminating harm we do what is right and we also
reduce waste thereby discharging our professional duty
to “do no harm” and to be “good stewards of finite
resources.
Primum non nocere Cura te ipsa