Transcript Clinical Decision Support: A Practical Guide to Developing

Optimizing Local Clinical Decision Support to Address National
Hospital Improvement Imperatives;
Early Efforts Toward a Scalable Collaborative
HIMSS Virtual Conference
and Exhibition
11/19/08
Session #107B
Jerry Osheroff
John Chuo
Anwar Sirajuddin
Donna Currie
Panel Desired Outcome/Agenda
©2008 Jerome Osheroff
• Desired Outcome: Attendees learn about
collaborative processes and strategies that can
enhance their CDS efforts and outcomes
• Agenda:
1. Describe effort to build and leverage a collaborative
to support CDS-enabled improvements in hospital
clinical imperatives
2. Hear from 3 of 6 organizations participating in the
collaborative: why they joined, what they are doing,
benefits in progress
3. 20 minutes of Q&A: to help ensure that attendees
realize desired outcome
Background on New HIMSS
Collaborative CDS Initiative, and
Implications for Attendees
Jerome A. Osheroff, MD, FACP, FACMI
Chief Clinical Informatics Officer, Thomson Reuters
©2008 Jerome Osheroff
Adjunct Assistant Professor, University of Pennsylvania
Collaboration Starting Point: Shared Need
and Opportunity to Improve Key Outcome
• VTE is leading cause of preventable hospital death
• Substantial costs for hospitals; beginning 10/08 CMS
and other payers not reimbursing for this hospital
acquired complication (and others)
• CDS is powerful tool for improving outcomes, but
implementations are complex and often problematic
©2008 Jerome Osheroff
• HIMSS, SI, others have collaborated successfully on
CDS best implementation practices
• Can we build on this success and create a dialog to
accelerate local CDS-enhanced performance
improvements?
©2008 Jerome Osheroff
Backdrop: Guides for Improving Care Processes
and Outcomes with CDS
www.himss.org/cdsguide
5
5
HIMSS/SI Initiative
• Goal: Develop scalable, validated guidance for provider
organizations to optimize CDS to drive measurable local
performance improvement on specific targets of high
local and national priority. (Initial focus=VTE)
• Formed CDS Task Force within HIMSS 8/08
• 6 clinical sites: Advocate, CHOP, HealthEast, Memorial
Hermann, Orlando, Texas Health Resources
©2008 Jerome Osheroff
• Other thought leader participants (CDS, performance
measurement, benefits realization) and Scottsdale
Institute
• HIMSS/SI will help with scaling
Deliverables: Near Term (HIMSS 4/09)
• Description of pilot site VTE CDS activities using
scalable templates and processes
• Models for successful practices’ for applying CDS to
VTE (given different CIS environments)
• Feedback about each site’s gap analysis based on
model practices, and plans based upon this analysis
©2008 Jerome Osheroff
• Ideas for scaling initiative to other topics and
organizations
Collaboration Process: Who and Why
• Who: Mostly CMIO types and related from prior
collaborative efforts; focus on PI, CDS and VTE
©2008 Jerome Osheroff
• Why: Implications for scaling and engaging others
(such as those in the audience)
Collaboration Process: How
• Calls, templates, wiki
– Calls: Agree on basic strategies (project scope,
deliverables, etc.)
– Templates: Standard ways of describing activities
– Wiki: Meeting schedules, develop talks
©2008 Jerome Osheroff
• Takes time to re-orient workflow to make best use of
wiki
• Have used a few wikis/Web 2.0 over various
collaborations; inevitable technical bumps in the road
Overall Lessons Learned (So Far)
• Does seem to be a need for and value from such
collaborations (see talks to come)
• Shared goal and early sense of value driving continued
participation
©2008 Jerome Osheroff
• Need skilled staff to support collaboration process:
agenda, minutes, documents, wiki management
(HIMSS providing this)
• Time management a challenge (as always) - few
hours/week/organization
So What?
Implications for You and Your Organization
• This work is manifestation of evolution toward “mass
collaboration” in society (see Tapscott’s Wikinomics)
• Such best practice syntheses and collaborations can
accelerate local efforts; consider jumping into the
HIMSS fray and/or beginning your own collaborative
©2008 Jerome Osheroff
• Engaging in HIMSS CDS efforts
– Review CDS guides and keep an eye on the CDS parts of
HIMSS website (e.g. www.himss.org/cdsguides)
– Send an email to David Collins ([email protected]) if you’re
interested in exploring participation in the VTE collaborative
when it scales later next year
About Memorial Hermann
About Memorial Hermann
• Memorial Hermann-TMC was recognized
as one of 100 U.S. hospitals to make the
greatest progress in improving hospitalwide performance over five consecutive
years (2001-2005).
• For the 18th consecutive year, Memorial
Hermann TIRR ranks among America’s
Best Hospitals in the U.S. News & World
Report magazine’s annual survey.
• Memorial Hermann Named A "Most Wired"
Memorial Hermann Fact Sheet
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Total hospitals: 14
• Acute care: 9
• Children’s: 1
• Heart & Vascular: 3
• Rehabilitation: 1
Regional affiliates: 21
Managed acute care hospitals: 4
Sports Medicine & Rehabilitation Centers: 27
Retirement/nursing center: 1
Home health agency: 1
Annual emergency visits: 323,258
Annual deliveries: 21,536
Annual Life Flight air ambulance missions: 3,185
Employees: 19,012
Beds (licensed): 3,286
Medical staff members: 4,194
Residency programs: 26 | Fellowship programs: 48
Physicians in training: 1,324 (physicians and fellows)
Annual community benefit: $229,152,000
Memorial Hermann IT Fact Sheet
 22,000+ devices connected to the enterprise network from 80 different locations
 4 data centers, 1 mainframe, 43 mid-range UNIX/VMS systems, 1000+ servers, 16,836 PCs, 5,023
printers, 561 scanners, 21 handheld devices
 22,400+ calls per month to the Support Center
 264,000 sessions per month on memorialhermann.org
 24,600+ Exchange Email user accounts
 12 million spam emails & 91,000+ network attacks blocked each month
 173 formal assigned projects
 56 telephone systems – 28,000+ phones, 8,000+ pagers, 2,434 Spectralink phones
 5.69 million unique person records in CARE4 with 17,654 total users
 9,289 unique physician users to CARE4 monthly
 300+ applications supported by ISD
 365 FTEs in the ISD organization and 106 located in the hospitals (471 total)
 $20M capital budget and $59.5M expense budget in FY 09
 3000+ Information Security access requests per month
VTE & Clinical Decision
Support
Why are we participating?
• Part of first collaborative
effort
• Experiences and best
practices
• Useful tool for CDS
implementers
Next Collaborative Effort
• Utilize principles from new guide
• Create scalable CDS models of best
practice for different clinical
conditions/diagnoses
• We almost missed the bus!
Why VTE?
• Significant cause of morbidity and
mortality despite the availability of effective
therapies for prophylaxis
• Non-reimbursement from CMS/other
payers
Challenges to collaboration
• Getting everyone together at the same
time!
• Using Wiki
• Overcame these challenges!
Our Objectives
• Learn the different processes
• Identify what would best work for us
• Share our experiences
• Outcomes
What are we doing today?
Memorial Hermann
•
•
•
•
•
Using online tool within EMR
Risk Assessment
Recommended Prophylaxis
All inpatients greater than 17 yrs
Physician completes risk assessment and
orders prophylaxis
• Nursing monitors compliance
What are we doing today?
Other Organizations
• Different organizations are quite different:
– VTE prevention initiative
– CIS implementation
– Different process
One single goal
Optimizing CDS to Prevent and Treat
VTE
Key Take Away Points
• Our process is more physician driven
• VTE Outcome Metrics
Next Steps for Project
• Continue to work on identifying core CDS
elements that can be scalable across
different organizations
• Identify best practices
• Implement best practices
• Do these best practices work?
Children’s Hospital of Philadelphia
Clinical Decision Support for
VTE Prevention
About CHOP
•
•
•
•
430 beds
34 sub specialties
Nearly 50 Sites within the CHOP network
HIS:
- Sunrise Clinical Manager, Manual access database for reporting (CS
Stars safetyNET)
- EPIC
• CDS related governance overview
- Executive Council of the medical staff --- ECMS IS --- CDS subcommittee
For the 6th consecutive year, CHOP ranks as the number one
Children’s Hospital in the U.S. News & World Report magazine’s
annual survey.
The Children's Hospital of Philadelphia
Why is VTE prevention Important?
• Hospital acquired VTE is potentially life-threatening and may;
- Prolong length of hospital stay
- Require invasive treatment
- Result in chronic disability, the need for follow-up care and long-term
anticoagulation
• Overall risk of VTE in children is much lower than in adults, but
children often have multiple risk factors
• Cases of hospital acquired VTE have occurred at CHOP
The Children's Hospital of Philadelphia
Improvement Aim
• To reduce the potential for harm through the
use of mechanical and chemoprophylaxis by
increasing the compliance with the clinical
practice guidelines to > 90% by February 2009.
The Children's Hospital of Philadelphia
Key Questions
1.
Are we assessing everyone who is at risk?
2.
Are we treating everyone who is at risk?
3.
Has treatment reduced the incidence of VTE?
The Children's Hospital of Philadelphia
OUTCOME MATRIX
ALL ADMISSIONS
Assessed with Risk score tool
At risk
Not at risk
+ VTE
Prophylaxis
+ VTE
- VTE
Not assessed with Risk score tool
- VTE
- VTE
No Prophylaxis
+ VTE
+ VTE
- VTE
The Children's Hospital of Philadelphia
MAXIMIZE
these groups
MINIMIZE
these groups
WHAT IS RISK and can it be stratified?
Age Distribution of Patients with Venous
Thrombosis at CHOP
40
Number of Patients
35
30
After the newborn
period, an increase
is seen at approx.
14 yrs
25
20
15
N=185
10
5
0
<1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Age (Years)
The Children's Hospital of Philadelphia
16
17
18
19
20 >21
Underlying Medical Conditions in Children with
Thrombosis
Diabetes
Sickle Cell Disease
Trauma
Renal
Infection
None
Other
Cystic Fibrosis
Cardiac
Dehydration
Prematurity
This does not include the presence of a central venous catheter, which is the single greatest
risk factor for thrombosis - found in ~40% of children over age 1 with a DVT
The Children's Hospital of Philadelphia
RISK assessment tool
At RISK patients are > 14 years old and have one or
more of the listed conditions
At HIGH RISK patients are > 14 years old, immobile,
and have one or more additional conditions
•
Immobility
•
Obesity
•
Orthopedic surgery
•
•
Spinal cord injury
•
Major trauma or trauma to the lower
extremities
•
Previous history of thrombosis / VTE
Disorder associated with thrombosis
including but not limited to;
Inflammatory Bowel Disease,
Nephrotic Syndrome, Sickle Cell
Disease, Lupus and Diabetic KetoAcidosis
•
Pregnancy
•
•
Central venous catheter in the lower
extremity
Taking estrogen containing medications
•
•
Acquired or inherited thrombophilia
Acute infection
•
•
Burns
Family history of thrombosis in a first
degree relative < 40 years of age
The Children's Hospital of Philadelphia
RISK assessment tool - execution
Defining
risk
groups
PAPER Retrospective
chart review,
consensus,
adult literature
Execution
Risk assessment tool is
incorporated into the nursing
admission intake paperwork
How well Outcome
executed is
the tool?
Are paper forms
of the tool readily
available?
Do people know
how to use it?
Is there
accountability?
TECH
Work underway for: Point of
care reminders in selected
ordersets to remind
prescribers to assess risk.
i.e. preop ordersets. The tool
itself will be computerized.
The Children's Hospital of Philadelphia
How many
risk
assessments
are done?
OUTCOME MATRIX
ALL ADMISSIONS
Assessed with Risk score tool
At risk
Not at risk
+ VTE
Prophylaxis
+ VTE
- VTE
Not assessed with Risk score tool
- VTE
No Prophylaxis
+ VTE
- VTE
The Children's Hospital of Philadelphia
+ VTE
- VTE
WHAT IS TREATMENT and can it be proven?
The Challenge of Treatment
• Treatment must …
1. Impact outcome (less VTE for at risk patients)
2. Safe
3. Determined by ….
• Research
• Published standards - from adult literature
• Consensus – Anticoagulant workgroup
• Randomized trials requires large numbers
Venograms too invasive, U/S technology and
MRI not good enough
The Children's Hospital of Philadelphia
Treatment Guideline
• A Patient who is at RISK or HIGH RISK receives one or more forms of
mechanical thromboprophylaxis*
• Early ambulation within 12 hours
• Pneumatic compression device*
• Graduated compression “antiembolic” stockings (Teds)
* Mechanical prophylaxis is not used if acute VTE is suspected
• A patient who is at HIGH RISK receives one or more forms of mechanical
prophylaxis and may also receive anticoagulant thromboprophylaxis with
low molecular weight heparin (Enoxaparin) or unfractionated heparin
The Children's Hospital of Philadelphia
Treatment - execution
Define
treatment
Execution
How well
executed
is the
process?
Outcome
PAPER Literature
review,
consensus,
adult
protocols
TECH
- VTE prophylaxis
order set
containing
mechanical and
chemoprophylaxis
- Nursing staff has
the ability to
initiate mech.
prophylaxis
Ordersets
awareness?
How often is prophylaxis
initiated?
Ordersets
usability?
What % of those
assessed to be at risk
received prophylaxis and
how do their VTE rates
compare with the other
groups?
The Children's Hospital of Philadelphia
Other key Measures for chemoprophylaxis
•Compliance with …
-Baseline laboratory monitoring (enoxaparin,
heparin, warfarin)
-Dosage adjustment algorithms and
monitoring guidelines (enoxaparin, heparin,
warfarin)
-Discharge Education and scheduled followup appointment (enoxaparin, warfarin)
The Children's Hospital of Philadelphia
The Children's Hospital of Philadelphia
OUTCOME MATRIX
= obtain from querying Hospital Information System
ALL ADMISSIONS
Assessed with Risk score tool
At risk
Not at risk
+VTE
Prophylaxis
+VTE
-VTE
Not assessed with Risk score tool
-VTE
No Prophylaxis
+VTE
-VTE
The Children's Hospital of Philadelphia
+VTE
-VTE
Lessons learned from workgroup
• We share the same challenges and have common strategies for overcoming the hurdles of
technology, culture, and clinical evidence.
• Risk assessment
-
We all stratify risk, some calculating a score.
Initial assessment - some done by nursing, some by physicians (about 50/50)
All working on electronic tool, but most using paper tools
Electronic alert triggers seems to be a good idea as reminder systems
• VTE Prophylaxis
- Order sets is the most popular way to group VTE related orders
- Some embed the order sets into existing ordersets associated with high risk populations
- Prompts to reminder clinicians to order VTE prophylaxis is a more difficult task than dose guidelines.
• VTE prophylaxis Complication prevention and outcome measures
- Compliance with guideline is a commonly measured metric among our group
- Prompts for appropriate labs is doable (i.e. INR)
• Improvement is temporally related to project activities and tapers off between activities.
• To hard-code practices into daily routine, it is necessary to insert continuous prompts at
the point of care
• Buy-in from and a sense of responsibility by front-line clinicians positively impacts the use
of VTE prophylaxis
• Identifying stakeholders is critical for success
The Children's Hospital of Philadelphia
Anticoagulant Workgroup
• Leslie Raffini, MD, Hematology
• Marilyn Blumenstein, RN, MSN
• Robert J. Mullen, PharmD,
CQPS
• Maria Mihalko, RN, MSN,
• Tara Trimarchi, RN, Chair of
CDS Committee
• Catherine Manno, MD
• Donna Schilling, BS, RT
• Jackie Evans, MD
• Jack Rome, MD (consultant)
• Daniela Davis, MD
• Kathryn Roberts, RN, MSN,
CRNP
• Amy Gallagher, PharmD, MS
• Lori Kramer, RN, MSN
• Sarah Erush, PharmD, BCPS
The Children's Hospital of Philadelphia
Venous Thromboembolism
(VTE)
DONNA CURRIE, MSN, RN
DIRECTOR, CLINICAL OUTCOMES
ADVOCATE HEALTH CARE
ADVOCATE HEALTH CARE



Faith based health care system
Largest fully integrated not-for-profit health care
delivery system in metropolitan Chicago
Organization
◦
◦
◦
◦
◦



8 hospitals (soon to be 9)
3,500 beds
Home Health Care
3 large Physician Groups
More than 200 sites of care.
25,000 employed associates
More than 4,600 affiliated physicians
Hybrid Medical Record
WHY WE PARTICIPATE
IN THE CDS PROJECT

To learn from others

To share information with other participants

To share information broadly

To identify opportunities for creative solutions
to common challenges
APPROACH TO VTE
PROPHYLAXIS








Six Sigma approach at 3/8 sites
System level interdisciplinary team
EMR site liaison participation
ACCP Guidelines
Assessment guidelines
Electronic assessment tool
Prophylaxis Guidelines
Local CDS committee
◦ VTE Alert #1 – MD alert for orders based on score
◦ VTE Alert #2 – Proposed alert for adequacy of
prophylaxis
VTE PATIENT RISK ASSESSMENT
GUIDELINES
Purpose
Provide a consistent process for assessing the risk of
VTE in medical and/or surgical patients.
Assessment Guidelines
- Within 8 hours of admission to the hospital
- With a change in level of care
- Post operatively for inpatient surgical patients
- Pre operatively for all surgical patients
- Notify the MD of the VTE risk assessment results,
suggested VTE prophylaxis and obtain orders
within 12 hours of admission to the hospital.
VTE PROPHYLAXIS GUIDELINES
PLEASE NOTE: These are GUIDELINES for the Prophylaxis of VTE. This is
NOT an order set. Please contact the patient’s physician for orders

VTE Prophylaxis Guidelines
• DO NOT USE ASPIRIN AS PROPHYLAXIS AGAINST VTE
• Discharge with LMWH for high risk general surgery patients; major cancer surgery
• Avoid thromboprophylaxis in vascular surgery patients with no additional risk factors
General
Low
(1)
• Early & Persistent mobilization
Moderate (2)
• Enoxaparin (Lovenox) 40 mg (If CrCI< 30mL/min = 30 mg) SubQ every 24 hours
OR
• Fondaparinux (Arixtra) 2.5 mg SubQ daily † (Do not use if CrCl< 30 mL/min or weight< 50 kg) OR
• Heparin, 5,000 Units SubQ every 12 hours
CONSIDER ADDING§
• GCS and/or IPC in addition to one of the above or if high risk of bleeding
High
• Enoxaparin (Lovenox) 40 mg (If CrCI< 30mL/min = 30 mg) SubQ every 24 hours
OR
• Fondaparinux (Arixtra) 2.5mg SubQ daily† (Do not use if CrCl< 30 mL/min or weight< 50 kg) OR
• Heparin, 5,000 Units SubQ every 8 hours
CONSIDER ADDING§
• GCS and/or IPC in addition to one of the above or if high risk of bleeding
(3-4)
Highest (> 5)
• Enoxaparin (Lovenox) 30 mg Sub Q every 12 hours (If CrCl< 30 mL/min = 30 mg daily)
OR
• Fondaparinux (Arixtra) 2.5mg SubQ daily† (Do not use if CrCl< 30 mL/min or weight< 50 kg)
OR
• Heparin 5000 units Sub Q every 8 hours
AND§
• GCS and/or IPC in addition to one of the above
† Approved for thromboprophylaxis in certain surgical patients
§ GCS = Graduated Compression Stockings
IPC = Intermittent pneumatic compression
MEASURES

Outcome Measure
◦ AHRQ measure
Actual post-op PE/DVT rate /expected post op PE/ DVT
rate

Compliance (Process) Measures
◦
◦
◦
◦
Risk assessment completion rates
Time from admission to VTE risk assessment
Time from admission to prophylaxis orders
Risk assessments by risk category




Low – 19%
Moderate – 14%
High – 28%
Highest – 39%
DRILL DOWN AT ONE SITE
Overall 64% medical vs.
36% surgical
 65% of patients scored
High or Highest Risk
 30% of all DVT/PE cases
reviewed grouped into a
circulatory MDC
◦ ¼ of these were bypass
w/o cath
◦ 83% did not receive
optimal prophylaxis

Pareto Chart of MDC
40
100
Count
60
20
40
10
20
0
0
l
.
.
.
y
s
C
g
s.
m
ta
m
m
m
ar
ou
ru
se
te
le
MD
in
st e
ste
st e
ys
ea
e
ea
fD
ke
Sy
Sy
Sy
Ur
is
Pr
yS
an
os
to
y
l
s
D
e
y
r
t
&
c
r
r
u
o
cu
ic
tiv
cu
to
l ia
vo
ey
f fe
at
us
si t
es
bi
ub
ra
er
ul
dn
cE
ra
ig
pi
to
-M
N
,S
rc
Ki
xi
rs
s-D
es
ep
in
Pa
Ci
To
of
rs
H
er
-R
Sk
sde
&
&
e
s
s
r
r
d
s
s
s
r
e
us
rd
er
er
er
iso
er
ng
rd
iso
io
rd
so
rd
D
rd
ni
rd
D
ct
so
i so
Di
so
&
fe
is o
so
iso
D
Di
&
Di
s&
D
es
In
Di
Po
&
s
e
s
&
,
&
s
e
&
s
a
&
s
a
e
es
as
es
se
es
se
ri
se
es
as
se
as
Di
as
ju
ea
as
Di
i se
Di
se
is
In
ise
se
D
Di
D
D
Di
s te
Sy
.
m
Count
12
6
6
4
3
2
2
2
1
1
1
Percent
Cum %
30.0
30.0
15.0
45.0
15.0
60.0
10.0
70.0
7.5
77.5
5.0
82.5
5.0
87.5
5.0
92.5
2.5
95.0
2.5
97.5
2.5
100.0
Percent
80
30
PROGRAM ANALYSIS

Advantages
◦ Standardized
 Approach to VTE prophylaxis
 Tools
 Measures

Opportunities
◦ Improve adequacy of prophylaxis
◦ Improve physician participation in the process
ANALYSIS OF CDS
PARTICPATION

Collaborative and iterative process – a
different way to work together

Learning from multiple, diverse
organizations

Develop creative, “Best Practice” processes

Share information to continually improve
practices
Topics for Discussion
• How are you addressing these issues (applying CDS
toward improvement priorities)?
• What are greatest CDS challenges, successes,
needs?
• Role for such a collaborative in your efforts?
©2008 Jerome Osheroff
• Other thoughts/questions?
More Information--Chair - HIMSS CDS Task Force
Jerry Osheroff, MD
[email protected]
©2008 Jerome Osheroff
HIMSS Staff Liaison
David Collins
Director, Healthcare Information Systems
[email protected]