Download Performance Improvement and Taxonomy

9/22/2015
Performance Improvement
and Taxonomy: Trauma
Taxonomy as a PI Tool
Annual Trauma Center Association of
America Conference
Donald H Jenkins, MD FACS
Professor of Surgery and Director of Trauma
Division of Trauma, Critical Care and Emergency General Surgery
Saint Marys Hospital, Rochester MN
September 2015
Disclosures
• None
Acknowledgements
•
•
•
•
•
•
•
Carol Immermann
Jorie Klein
Chris Cribari
Blaine Enderson
Glen Tinkhoff
Ronald Stewart
Eric Epley
•
•
•
•
•
•
•
Kathie Martin
ACS COT PIPS
STN/TOPIC
Terri Elsbernd
Brenda Schiltz
Deb Horsman
Chris Ballard
1
9/22/2015
What is most important with
respect to assessing trauma center
care? A range of view points.
VRC Processes
TQIP Outcomes
Current Verification
Good (risk
adjusted)
Outcomes
Good VRC
Processes
and
Structure
✔
Poor VRC
Processes
and
Structure
✖
Poor (risk
adjusted)
Outcomes
✔
✖
How to Incorporate Outcomes into
Verification?
Good (risk
adjusted)
Outcomes
Poor (risk
adjusted)
Outcomes
Good VRC
Processes
and
Structure
✔
✔?
Poor VRC
Processes
and
Structure
✖?
✖
2
9/22/2015
Model Best
Practice Site
Study and
disseminate
approach to
care
+ Implement
Future VRC Approach?Verifiy
approved COT PIPS
Good VRC
Processes
and
Structure
Poor VRC
Processes
and
Structure
Good (risk
adjusted)
Outcomes
✔
Plan
Poor TQIP Best Practices
RA Outcomes
✔
✖
✖ ?✔
? Verify + Implement approved COT PIPS
Plan TQIP Best Practices
Not verified
Implement approved COT PIPS Plan
TQIP Best Practices
Culture of Safety Taxonomy
Objectives
• Origin of the culture of safety taxonomy
• Goals of Implementation
• 5 Nodes of the Taxonomy
•
•
•
•
•
Impact
Type
Domain
Cause
Mitigation/prevention (event resolution)
• Benefits of adopting this new taxonomy
The National Quality Forum Taxonomy has
been recommended by the ACS COT PIPS
Committee and the ACS VRC as best practice
and should be implemented in trauma
performance improvement programs. This
taxonomy should be implemented now and will
be a criterion (required) in the next version of
the Resources for Optimal Care of the Injured
December 2013
3
9/22/2015
National Quality Forum (NQF)
Taxonomy Step-By-Step
Implementation
• Do PI the way you have always done PI
• Events act as triggers for case review:
• Deaths
• Other non-discretionary events (i.e., specific complications [i.e., NTDS
complications])
• Discretionary events
• Essential criteria that are integral to PI are used
• Classify the relevant factors for the event using the NQF taxonomy
- Define cut offs for primary review, secondary review, tertiary review
Define cutoffs for track/trend versus action plan development
-
• Develop computerized application to enhance ease of use
• Import NTDS complications as baseline sentinel events
• Allow users to add additional sentinel event types
Taxonomy
• Building blocks
• Common definitions and classifications
• Unambiguous and translatable terminology
• Scope
•
•
•
•
Comprehensive classification tool
Applicable to all health care delivery settings
Includes multiple levels of patient harm
Addresses sentinel or serious events, adverse
events, no-harm events, near misses or close
calls, and potential events
11
12
4
9/22/2015
Taxonomy
(Ivatury et al. JT, Feb 2008)
• Impact: Outcome or effect of event
• Type: Processes that were faulty
• Domain: Setting where incident occurred or
phase of care
• Cause/Factors: System and human factors
and agents that led to incident
• Prevention and mitigation: Universal,
selected or indicated, an action plan
13
Framework of the Taxonomy
Severity of harm (e.g. TJC sentinel events reports, NCC MERP)
Type of health care service provided I. Impact
Type of individual involved (physician, nurse, etc.) and type of setting (hospital, clinic, etc.)
II. Type
‐Overuse, underuse, misuse
‐Active & latent failures
‐Negligence
III. Domain
IV. Cause
Primary Classifications Further
Defined
1. Impact: the outcomes or effects of medical error
and systems failure, commonly referred to as
harm to the patient.
2. Type: the implied or visible processes that were
faulty or failed.
3. Domain: the characteristics of the setting in
which an incident occurred and the type of
individuals involved.
4. Cause: the factors and agents that led to an
incident.
5. Prevention and Mitigation: the measures taken
or proposed to reduce the incidence and effects
of adverse occurrences.
5
9/22/2015
Classification: Impact
Medical
Non-Medical
Psychological
Physical
Legal
Social
I. No harm/no
detectable
harm
II. No
detectable
harm
I. No harm/no
detectable
harm
II. No
detectable
harm
III. Mild
temporary
harm
IV. Mild
permanent
harm
III. Mild
temporary
harm
IV. Mild
permanent
harm
Extremely
satisfied
Satisfied
V. Moderate
temporary
harm
VI. Moderate
permanent
harm
V. Moderate
temporary
harm
VI. Moderate
permanent
harm
Neutral
Dissatisfied
VII. Severe
temporary
harm
VIII. Severe
permanent
harm
VII. Severe
temporary
harm
VIII. Severe
permanent
harm
Extremely
dissatisfied
IX. Profound
mental harm
Economic
Patient/Family
Satisfaction
IX. Death
Differentiating Levels of Harm
• None – patient outcome is not symptomatic or no
symptoms detected and no treatment is required (I. & II.
Impact)
• Minimal – patient outcome is symptomatic, symptoms are
mild, loss of function or harm is minimal or intermediate
but short term, and no or minimal intervention (e.g., extra
observation, investigation, review or minor treatment) is
required (III. & IV. Impact)
• Moderate – patient outcome is symptomatic, requiring
intervention (e.g., additional operative procedure;
additional therapeutic treatment), an increased length of
stay, or causing permanent or long term harm or loss of
function (V. & VI. Impact)
Differentiating Levels of Harm
• Severe – patient outcome is symptomatic, requiring lifesaving intervention or major surgical/medical
intervention, shortening life expectancy or causing major
permanent or long term harm or loss of function (VII. &
VIII. Impact)
• Death – on balance of probabilities, death was caused or
brought forward in the short term by the incident (IX.
Impact)
6
9/22/2015
IMPACT
The outcome or effects of a medical error and systems failure, commonly
referred to as harm to the patient
Physical
1. No Harm & No Undetectable Harm-Sufficient information or able to determine that no
harm occurred
2. No Detectable Harm-Insufficient information or unable to determine any harm
3. Minimal-Temporary Harm- Requires little or no intervention
4. Minimal Permanent Harm-Requires initial but not prolonged intervention
5. Moderate-Temporary Harm- Requires initial but not prolonged hospitalization
6. Moderate-Permanent-Harm-Requires intensive but not prolonged hospitalization
7. Severe-Temporary Harm-Requires intervention necessary to sustain life but not
prolonged hospitalization
8. Severe-Permanent Harm- Requires intervention necessary to sustain life and
prolonged hospitalization, long-term care, or hospice
9. Death
Classification: Type
Patient
Management
Communication
Clinical
Management
Inaccurate &
incomplete
information
Questionable
delegation
Questionable
advice or
interpretation
Questionable
tracking or
follow-up
I. Correct
diagnosis,
questionable
intervention
I. Correct
procedure with
complication
II. Correct
procedure
incorrectly
performed
I. Correct
prognosis
Questionable
consent process
Questionable
referral or
consultation
II. Inaccurate
diagnosis
III. Correct
procedure but
untimely
IV. Omission of
essential
procedure
II. Incorrect
prognosis
Questionable
disclosure
process
Questionable use
of resources
III. Incomplete
diagnosis
V. Procedure
contraindicated
VI. Procedure not
indicated
III. Incomplete
prognosis
IV. Questionable
diagnosis
VII. Questionable
procedure
VIII. Wrong
patient
IV. Questionable
prognosis
Pre-Intervention
Questionable
documentation
Intervention
Post-Intervention
Classification: Domain
Period
Setting
Hospital
Non-Hospital
Staff
Physicians
Date
Nurses
Patient
Therapists
Others
Physical
therapist
Health
professions
student
Target
Age
Diagnostic
Gender
Therapeutic
Emergency room
Ambulatory care
Practitioner]s
Office
Subacute care
Skilled nursing
care facility
Ambulatory Care
Clinic
Month
Resident
Licensed
practical nurse
Occupational
therapist
Pharmacist
Diagnosis
Rehabilitative
Diagnostic
procedures
Clinical
laboratory
Nursing Home
Day
Attending
Registered nurse
Speech therapist
Pharmacy
technician
Coexisting
Conditions
Preventive
Home Care
Holiday
Dentist
Nurse
practitioner
Radiation
technician
Duration of
Disease
Palliative
Podiatrist
Optometrist
Socioeconomic
Status
Research
Physician
assistant
Other
Education
Cosmetic
Other
Other
Rehabilitation
Hospice
Other
Mental health
Pharmacy
Hospice
Rehabilitation
Facility
Mental health
Facility
Year
Time
Intern
Nurse’s aide
Other Facility
7
9/22/2015
Classification: Cause
Human (actual or near
misses)
Structure/Process
Organizational
Technical
External to
organization
Management
Facilities
Organizational
culture
Protocols/
procedures
External
Patient
Practitioners
Patient factors
Other
External
Skill-based
Negligence
Recklessness
Rule-based
Transfer of
knowledge
Knowledge-based
Unclassifiable
Classification:
Prevention (P) & Mitigation (M)
aka Action Plan
Universal
Improve the accuracy
of patient
identification (P)
Improve the
effectiveness of
communication
among caregivers (P)
Improve the
effectiveness of
clinical alarm systems
(P)
Reduce the risk of
healthcare-acquired
infections (M)
Selective
Eliminate wrong-side,
wrong-site, wrongprocedure surgery
(M)
Indicated
Improve the safety of
using high-alert
medications (P)
Improve the safety of
using infusion pumps
(P)
TJC Taxonomy Via Software
• Advantages
• Ease of use
• Improved data collection
• Improved data collation
• Disadvantages
• Development time
• Distribution
• Training
8
9/22/2015
Why Do This?
• We will be able to PI our PI
• Benchmark our PI
• Consider risk stratifying TQIP reportable
events (complications) (?using Charlson
Index?)
• Focus PI on TQIP AE’s?
• Focus PI on TQIP AE’s that are outliers on
benchmarking report?
• Focus PI on Hospital QA initiatives and
customize registry?
• Chpt 16 says to align trauma PI with hospital
quality programs
ACSCOT Update
• Enhance the interaction of PIPS with NTDS, NTDB, VRC
and TQIP
• Definitions of NQF taxonomy are not ‘traumafied’ and are
incomplete from NTDB and TQIP perspective (this is
being fixed)
• Many NTDB and TQIP adverse events have elements
that are not defined in the NQF taxonomy (this is being
fixed)
• Evaluate best practices and advise low performing
centers on these
• Commitment to ‘fix’ this as of 8 Jan 15 COT Executive
Meeting
Trauma PIPS Levels of Review
Primary Review
Opportunity for Improvement/Validation
Adverse Event/
Audit Filter
Review
Secondary Review
Issues
Elevated to
Hospital PIC
Monthly
Tertiary Review
Trauma
Peer Review
(Monthly)
Trauma
Committee
(Monthly)
Trauma M&M
(Weekly)
Actions
Education
Counseling
Track/Trend
Guideline
Development
PIPS Team Project
9
9/22/2015
Benchmark Comparison with
NTDB
Compare your trauma hospital data with national data
Examples:
Examples:
• Patient
Demographics
• Hospital
demographics
• Survivors vs. nonsurvivors:
• LOS
• mean ISS & ICU
days
• Age
• Blunt vs. penetrating
percentages
• ISS by age group
• Mortality rates
• Mortality by ISS
• ED disposition
• Hospital disposition
• ISS and hospital charge
• Mechanism of injury and
restraint usage
• ISS with LOS
28
Benchmarks and
Measurements:
Outcome Data
Report Examples:
• Functional status on discharge (FIM Scores)
• Results of patient satisfaction surveys
• Complication rates
• Compliance with practice management guidelines
• Mortality and morbidity
• Severity-adjusted mortality and morbidity
• Unplanned return to OR
• Unplanned upgrade to an intensive care unit
• Unplanned hospital readmission
• Surgical wound infections
• Organ donation activity
29
Trauma Quality Improvement Project
[TQIP] Risk Adjusted Benchmarking
• Required at Level I, II, and III centers
• Reduce variability in trauma process/outcomes/cost
• A trauma PIPS program for evaluating risk adjusted
performance and benchmarking
• Goals:
• Reduce variability in trauma care thus improving
outcomes and decreasing costs
• Develop data elements to measure processes of care
• Standardize care management via trauma centers
nationally
• Implement uniform defined audit filters and universally
accepted data definitions
• Develop model PI plan and program
30
10
9/22/2015
TJC Taxonomy Implementation
• Do PI the way you have always done PI
• Events act as triggers for case review:
• Deaths
• Other non-discretionary events (i.e., specific complications [i.e., NTDS
complications])
• Discretionary events
• Essential criteria that are integral to PI are used
• Classify the relevant factors for the event using the TJC taxonomy
- Define cut offs for primary review, secondary review, tertiary review
Define cutoffs for track/trend versus action plan development
-
• Develop computerized application to enhance ease of use
• Import NTDS complications as baseline sentinel events
• Allow users to add additional sentinel event types
Questions?
11