Interface Terminology - Vanderbilt University Medical Center

Download Report

Transcript Interface Terminology - Vanderbilt University Medical Center

Clinical Terminologies for
Research
S. Trent Rosenbloom, MD MPH
Assistant Professor, Department of Biomedical Informatics
Assistant Professor in Clinical Nursing, School of Nursing
Adjunct Faculty, Departments of Internal Medicine & Pediatrics
Vanderbilt University Medical Center, Nashville, TN, USA
Pithy Quote
Water, water, everywhere,
and all the boards did shrink;
Water, water, everywhere,
nor any drop to drink.
Samuel Taylor Coleridge, The Rime of the Ancient Mariner, 1798
Significance
There is a lot of patient data in clinical
documents (i.e., Water, water, everywhere)
Most data is difficult to extract for research
or automation of routine operational tasks
(i.e., nor any drop to drink)
Terminologies may improve the accessibility
of clinical data.
What is a Terminology?
In my research database:
– “Myocardial Infarction”
– “Heart Attack”
In another researcher’s database
– “Myocardial Infarct”
This leads to problems of ambiguity and
with interoperability
What is a Terminology?
The problem is not just theoretical:
– At our institution, weight is stored in at least
ten formats within the same EHR system
–
–
–
–
–
–
–
–
WizOrder (>3)
StarPanel
StarNotes
StarForms (?n)
Quill (can reuse data, user does not always opt for this)
Dictations
Research intake forms (?n)
Others?
How do I perform efficient HSR and QA?
What is a Terminology?
Morning Star
Evening Star
The second planet from the sun, having an average radius of 6,052 kilometers (3,761 miles), a
mass 0.815 times that of Earth, and a sidereal period of revolution about the sun of 224.7 days
at a mean distance of approximately 108.2 million kilometers (67.2 million miles).
What is a Terminology?
Physical Entity
The second planet from the sun, having an average radius of 6,052 kilometers (3,761 miles), a
mass 0.815 times that of Earth, and a sidereal period of revolution about the sun of 224.7 days
at a mean distance of approximately 108.2 million kilometers (67.2 million miles).
Representative Terms
Morning Star
Conceptual Experience
Evening Star
Venus
Adapted from Campbell, ‘Representing thoughts, words, and things in the UMLS’, 1998.
What is a Terminology?
Planets of the Solar System
inside
outside
Mercury
Jupiter
Venus
Saturn
Earth
Neptune
What is a Terminology?
Terminologies are formal representations
of entities and their interrelationships.
– Embodied as concepts and terms
Concepts are universal definitional meanings
Terms are evocative words or phrases.
What is a Terminology?
Concept: Myocardial Infarction
CUI: C0027051
Semantic Type: Disease or Syndrome
Entity: Gross necrosis of the myocardium, as a result of interruption of the blood supply
to the area. (Dorland, 27th ed)
Representative Terms (synonyms):
Myocardial Infarction
Attack coronary
Cardiac infarction
Heart attack
Infarction of heart
MI
MI - Myocardial infarction
Myocardial Infarct
Myocardial infarction (disorder)
Myocardial infarction syndrome
myocardium; infarction
Adapted from the UMLS Metathesaurus.
More Specific Concepts (children):
Acute myocardial infarction
Old myocardial infarction
Microinfarct of heart
True posterior wall infarction
Aborted myocardial infarction
Other specified anterior myocardial infarction
Silent myocardial infarction
Subsequent myocardial infarction
Postoperative myocardial infarction
First myocardial infarction
Myocardial infarction with complication
Non-Q wave myocardial infarction
What is a Terminology?
Best terminology practices can help
improve research data in terms of:
Interoperability
Clarity
Consistency
Query-ability (accuracy and precision)
Manipulability
What is a Terminology?
Terminology Desiderata
Statement of purpose, scope, and comprehensiveness
Complete coverage of domain specific content
Use of concepts rather than terms, phrases and words (concept orientation)
Concepts do not change with time, view or use (concept consistency)
Concepts must evolve with change in knowledge
Concepts identified through nonsense identifiers (context-free identifier)
Representation of concept context consistently from multiple hierarchies
Concepts have single explicit formal definitions
Support for multiple levels of concept detail
Absence of or methods to identify duplication, ambiguity, and synonymy
Integration with other terminologies
Mapping to administrative terminologies
Bottom line:
comprehensively represents a medical domain with all
the concepts it contains, called Completeness
Adapted from Cimino, ‘Desiderata for controlled medical vocabularies in the twenty-first century’, 1998.
What is a Terminology?
Completeness achieved by one of two ways
Post-coordination - complex concepts from different
levels of detail are composed as needed from
fundamental concepts
(e.g., ‘chest pain’ composed from the concepts ‘chest’ and
‘pain’ when needed)
Pre-coordination - all levels of detail are modeled
with distinct concepts
(e.g., ‘chest pain’, ‘substernal chest pain’, and ‘crushing
substernal chest pain’ are all in the terminology)
What is a Terminology?
What is a Terminology?
Completeness measured by Coverage:
coverage calculated as the proportion of concepts
covered by a terminology
multiple studies: post-coordinated terminologies
generally have better coverage than precoordinated terminologies
What is a Terminology?
Consequences of post-coordination:
Inefficient post-coordination:
cumbersome
for
D5-46100 “too
01 Appendicitis,
NOS
M-41000complex
01 Acute inflammation,
NOS
01 Acute
problem
entry” G-A231
G-CO06 01 In
M-40000 01 inflammation, NOS
D5-46210 01 Acute appendicitis, NOS
G-A231 01 Acute
T-59200 01 Appendix, NOS
G-CO06 01 In
T-59200 01 Appendix, NOS
Table. Duplication due to compositionality: four ways to compose ‘Appendicitis’ in SNOMED, from the
CANON Group
Nonsensical Concepts
Concept duplication
What is a Terminology?
Terminologies come in different flavors
Interface
Terminology
Adapted from Chute, Layers of Terminology Processing, in ‘Desiderata for a Clinical Terminology Server’, 1999.
The Problem
My colleagues down the hall have a grant to
pay for data extraction from clinical records:
– “The patient has MR”
– “LAD disease was noted by the referring doc”
Data extraction usually isolated from patient
care episodes
Documentation often performed without
considering subsequent data extraction
Clinical Documentation
Encounters between patients and healthcare providers are
typically recorded in clinical documents.
clinic notes,
procedural narratives,
comprehensive history and physical reports,
periodic assessments,
progress notes, consultation notes,
hospital discharge summaries
Clinical documentation creates an objective record of
physical findings, medical reasoning, and delivered care
Many additional demands are imposed on clinical
documentation by third parties
Clinical Documentation
The federal government and major stakeholders in
healthcare emphasize the importance of rapidly
deploying electronic health record systems to all sites
of health care.
increase patient safety
reduce errors
improve efficiency
reduce costs
Capturing structured information from clinical notes
can improve healthcare and research, but remains a
major challenge.
Clinical Documentation
Medical documents have long served as legal treatises,
recounting the care and procedures that individual patients
receive for the purposes of potential future arbitration.
Insurers use the quality and quantity of documentation to
justify the level of reimbursement for given services.
Elements of medical records may be used as markers for
the quality of care provided to patients.
Clinical documentation content may be used as a substrate
for decision support systems.
Each of these forces defines additional components that must
be included in medical documents.
Clinical Documentation
Current state of the art
Paper based / Impermanent
Asynchronous
Redundant
Expensive
Illegible (or inefficient to transcribe)
Location-Bound
Limited functionality
Unchanged in 5000+ years
Clinical Documentation
Clinical Note circa 3000 BCE
Chief Complaint
History of Present Illness
Clinical Documentation
Past Medical History
Clinical Note circa 2000 CE
Family History
Social History
Review of Systems
Physical Exam
Clinical Documentation
To date, most CNCTs described in the literature
serve only isolated domains or single problems
Workup of febrile neonates,
Evaluation of providers who have had blood and body fluid
exposures,
Generation of progress notes in the neonatal and medical
intensive care units,
Enhancing compliance with standards of diabetes care,
Documenting pediatric care based on national guidelines,
Construction of obstetric databases
Documentation of endoscopy procedures
Only a few general purpose CNCT have been
described, although they may be increasing
Clinical Documentation
Clinical Documentation – current
Clinical Documentation
Real-time documentation using categorical
information is called Structured Entry
Structured Entry can improve patient care
by generating data in real time
Structured Entry can generate data to
support clinical research
– Requires an “Interface Terminology”
Background
A Structured Entry tool
Clinical Documentation
Structured entry of clinical notes currently poorly
adopted in healthcare
Reasons for poor adoption not well characterized
Inefficient to use
Complicated interfaces
Inflexible in face of unforeseen clinical problems
Difficulty finding what you want to say
The terminologies supporting structured entry
tools may play a significant role
Terminologies
There currently exist many terminologies
In 2003, the National Committee on Vital Health
and Statistics (NCVHS) recommended a subset of
existing terminologies as:
“uniform data standards for patient medical
record information (PMRI) and the electronic
exchange of such information”
Terminologies
PMRI standards:
– SNOMED CT (as licensed by the National Library of
Medicine) - for the exchange, aggregating, and analysis
of patient medical information.
– Logical observation Identifiers Names and Codes - for
the representation of individual laboratory tests
– Federal Drug Terminologies:
RxNorm;
The representations of the mechanism of action and
physiologic effect of drugs from NDF-RT;
Ingredient name, manufactured dosage form and package
type form the FDA
Terminologies
PMRI recommended mapping priorities:
– Priority 1: Terminologies previously designated
as HIPAA medical code sets:
CPT-4 (Current Procedural Terminology)
CDT (Current Dental Terminology)
ICD-9-CM (International Classification of Diseases –
Clinical Modification)
NDC (National Drug Codes)
Terminologies
PMRI recommended mapping priorities:
– Priority 2: Terminologies in common use as
enablers of important healthcare functions.
DSM-IV (diagnosis codes for mental disorders)
ISBT 128 (coding system for describing blood
products and tissues)
Medcin (codes for structured entry of clinical notes)
MedDRA (international code set for use by drug
regulatory agencies)
Nursing terminologies not otherwise included in
SNOMED CT
Terminologies
What about the UMLS?
(Unified Medical Language System)
– The UMLS is a terminology collection
– Concepts are unique and may fit the best
practices desiderata
– No formal relationships among concepts
present, per se
Terminologies
Using the UMLS:
– Semantics and relationships from source
terminologies lost (or worse, implied)
– May mix up different levels of detail from
different terminologies
– Can loose link with source terminology,
which can hinder maintenance.
Conclusion
Clinical terminologies
– are formal arrangements of terms, concepts
and their interrelationships
– can represent knowledge and data if
generated according to best practices
Challenges remain
– Getting data from narratives
– Creating flexible structured entry tools
– Picking the right terminology to fit the intended use
Acknowledgement
Assistance in developing this work:
Steven H. Brown
Kevin B. Johnson
Peter L. Elkin
Ted Speroff
Randolph A. Miller
Judy G. Ozbolt
supported by NIH 1 K22 LM08576-01
Discussion
Questions?
[email protected]
Thank You!