Transcript Finding and Using Knowledge Resources

Bringing Clinical Knowledge to the
Point of Care and to Patients
Jonathan Teich
Patti Abbott
Bill Hersh
Molly Mettler
Eduardo Ortiz
Marc Overhage
Nick Beard
Topics
Clinical decision support: examples,
value, best practices
Finding and using knowledge resources
Turning clinical goals into computer
interventions
CDS as a “team sport” – nurses,
patients, and more
Clinical Decision Support and the
Quality of Care
Jonathan Teich, MD, PhD
SVP and Chief Medical Officer
Healthvision
Asst. Prof. of Med., Harvard Univ.
“Providing clinicians or patients with
clinical knowledge and patientrelated information, intelligently
filtered and presented at appropriate
times, to enhance patient care”
The Goal
Patient Safety
Improved Quality
Improved Communication
Adherence to Best Practices
Improved Access to Knowledge/Evidence
Better Financial Performance
Substitution alerts
Consequent (corollary) orders
Impact studies - Excessive doses
2.5
% of orders
2
1.5
1
0.5
0
Pre-OE
Post-OE
Year 1
Year 2
Year 3
% of orders
Use of recommended H2 blocker
100
90
80
70
60
50
40
30
20
10
0
-7
-6
-5
-4
-3
-2
-1
0
1
Week
2
3
4
5
6
7
8
Ambulatory errors
Reported in 5% to 18% of encounters
1 per 20 patient-years, 40% very serious
Leapfrog and IOM new reports on ambulatory
errors
Allergies and DDI‘s, drug-food and drug-OTC
Drug monitoring, especially Coumadin
Poor protocol follow-through
Poor test follow-up
Between-visit condition changes
Patient misinterpretation
Ambulatory Clinical Decision Support
Drug allergy, DDI
Drug substitution and preconditions
Appropriate utilization of tests
Order sets and care plans (algorithms)
Asynchronous warnings and alerts
Health maintenance reminders
Prescribable education
Just-in-time reference information
Formulary Checking
www.himss.org/cdsworkbook
Opportunities for CDS
Six steps to successful CDS
Determine the CDS program’s goals and
clinical objectives
Catalog your information systems’
capabilities to help achieve those targets
Select the best interventions and
opportunities to achieve specific
interventions
Review and vet interventions with
stakeholders
Monitor how it is working
Turning Decision Support Goals into
Interventions
Patricia A. Abbott
Assistant Professor & Director of the
WHO/PAHO Center for Information
Systems in Nursing Care
Johns Hopkins University School of
Nursing
People – The Key to Success
CDS impact on behavior
Humans do not like cheese moved
Identify key stakeholders/opinion leaders
Interdisciplinary
Key Success Factors1
Deep executive support & belief in IT
History of successful clinical IT
Excellent communication to stakeholders
Involvement of key users & champions
Strong support from IT staff for problem
resolution – before, during and after
implementation
1. Osheroff, Pifer, Sittig, Jenders & Teich (2004). Clinical Decision Support Implementers Workbook.
Identifying the Achievable Goal
Goals determined in collaborative fashion
Evaluate current efforts
• Case management
• Process redesign
Identify fertile areas of importance
Consider goals individually and collectively
Focused, measurable, and realistic
From Goal to Action
High level Goal
Focused Goal
Clinical Goal
Clinical Action
Progra
m
Goals
Objectives
Objective Classes
Interventions
Parameters
Example
High-level goal / program:
Support Disease
Mgmt. Programs that measurably improve care & outcomes
Focused Goal:
Program
Diabetes Mellitus Disease Mgmt.
• Clinical Goal: Prevent Diabetic Retinopathy
– Clinical Action - Increase regular Ophthalmology f/u
– Clinical Action - Perform annual funduscopic exam
• Clinical Goal: Decrease complications of diabetic
neuropathy
– Clinical Action – Increase regular podiatry follow-up
• Clinical Goal: Prevent diabetic nephropathy
– Clinical Action – Increase regular measurement of
microalbumin
Osheroff, Pifer, Sittig, Jenders & Teich (2004). Clinical Decision Support Implementers Workbook.
Selecting Interventions – CDS Types
• Forms and templates (encounter
documentation)
• Relevant data presentation (flowsheets, CPM)
• Order sets
• Integrated guidelines (active guidelines)
• Reference information (links/Infobuttons)
• Reactive/unsolicited alerts (drug interactions)
Finding and Using Knowledge Resources
William Hersh, MD
Professor and Chair
Department of Medical Informatics & Clinical
Epidemiology
Oregon Health & Science University
[email protected]
The life-cycle of medical (scientific)
knowledge
Secondary
publications
Publish
Relinquish
copyright
Original
research
Revise
Public data
repository
Write up
results
Reject
Accept
Peer
review
Submit for
publication
The hierarchy of evidence
(Haynes, 2001)
Systems –
computerized
decision support
Synopses – evidencebased abstractions
Syntheses – systematic reviews
and evidence reports
Studies – original articles published in journals
Knowledge resources
Systems
Guidelines, rules
Synopses
Textbooks, compendia, guidelines
Syntheses
Databases of systematic reviews
Studies
MEDLINE
Journal
articles
Finding studies
pubmed.gov
www.bmj.com
Finding syntheses
www.cochrane.org
http://www.ncbi.nlm.nih.gov/
entrez/query.fcgi?db=Books
Finding synopses
www.guideline.gov
www.clinicalevidence.com
Challenges in knowledge generation,
access, and use
Generation – development of syntheses,
synopses, and systems for use at point
of care
Access – open access, interoperability,
licensing
Use – how to apply in the clinical setting
Using knowledge resources at the point
of care – some general pointers
Clinicians need the appropriate amount of the
right information at the right time
Systems and synopses at point of care
Syntheses and studies when have more time or
need more detail
Interesting research coming down the pike
Infobuttons – Cimino et al. and others
Maintaining user “context,” i.e., allowing user to drill
down to more information at a later point
What about different devices, e.g., PDAs and
tablets?
Each has benefits and drawbacks; ideal solution is
allow use of any
As with EHR, need standards for display, query,
rights management, etc.
CDS as a Team Sport:
The Role of Nurses, Allied Health
and Consumers
Molly Mettler
Senior VP
Healthwise
CDS that Supports
Interdisciplinary Care
Evidence Based Practice depends on coordination of all
providers, including nurses, allied health, social work, &
physicians
“Domino” Effect
Easy to use by busy clinicians
Fits into workflow
Use of guidelines & documentation as natural process
Focused, nimble, sensible, & vetted by
users/consumers
Patient-Centered Informatics
Include the patient
Facilitate patient/provider
conversations
Support shared-decision
making
Support patient followthrough
Information Therapy:
A Patient-Centered Solution
The prescription of :
 specific evidence-based medical information
 to a specific patient, caregiver, or consumer
 at just the right time to help them make a specific health
decision or behavior change
as part of the process of care
Diabetes Self-Management:
The Consumer’s POV
Diet
Exercise
Medications
Eye/Foot Care
Insulin Monitoring
Depression
Nerve damage
Etc., etc., etc.
Using a plate format for people with diabetes
The Ideal Application of CDS/CDA
From electronic
medical record… to
patient portal
Easy to Use ● Focused and Specific ● Documented