The Importance of Dynamism in the OpenVistA Model

Download Report

Transcript The Importance of Dynamism in the OpenVistA Model

The Importance of
Dynamism in the
OpenVistA Model
Chris Richardson & David Whitten
WorldVistA
http://www.worldvista.org
June 16, 2005
WorldVistA
• Nonprofit, 501(c)(3) public-benefit corporation
– Seeks to improve health worldwide by making medical software
more accessible, more affordable, and better.
• Members include
– Developers working, or who used to work, for U.S. DVA
– OpenVistA end-users
– Other organizations members or representatives
• Promote VistA Open-Source
– Future development coordination
• Including “localization” and “internationalization”…
– Change management & licensing
– Inter-institutional arrangements and coordination
– End-User and Developer Education & Training
2
What is VistA?
• Most integrated Health Information
Technology (HIT) system in the world
– Integrated outpatient and inpatient information with
local flexibility and innovation
– 100+ Modules
• Used by over 100,000 users worldwide
– Including 1,300+ VA facilities; in 20+ care networks
• 170+ Hospitals
• 650+ Outpatient Clinics
– 25+ Million Veterans - 20% some receiving care
in more than one facility/location
» Shows List of Sites with Last Treatment Date
» Displays CPRS Reports from VHA Facilities
• Home Care, Long-term Care Facilities, etc…
3
What is VistA?
• HIT built on standard, industrial strength technologies
– Leverages and supports Open Source software community
• Architecture
– Client
• GUI for critical and timely data access at all levels
• Terminal-interface available for power-users.
– Security
• Secure communications between client and servers
– Database
• From FOSS (GT.M) to commercial offerings (Caché)
• Platform managing millions of patient records for decades
• Blazingly-fast data stores
– Up to 20 times faster than RDBMS and SQL…
– Operating Systems and Hardware
4
• From “big iron” to commodity 32 and 64-bit hardware platforms
• From Windows to Unix/Linux servers
What is VistA?
• Multimedia Patient Records
– Text
– Computable data
– Handwritten and non-electronic data
• Via “scanning”
– Other multimedia data accessible…
• Radiology images
• EKG waveforms
• Other medical images
– pathology slides, endoscopic views, dermatology photos, …
• Video clips
– echocardiogram, cardiac cath films, barium swallows, gait analysis, …
• No double-entry of information
5
What is VistA?
• Simultaneous user accesses
– Remote access when needed
• No loss of patient information
• Contains “traditional chart components” such as:
–
–
–
–
–
–
–
6
Progress Notes
Orders
Procedure Reports
Discharge Summaries
Problem Lists
Lab Results
Vital Signs
What is VistA?
• Supports “other chart components” such as:
–
–
–
–
–
–
Annotated Diagrams - Stick Figures
Advanced Directives
Consent Forms
Flowsheets
Admission and Referral Forms
Reports from other medical facilities
• Augments chart components with:
–
–
–
–
7
Automated alerts for allergies, directives
Periodic Clinical Reminders
Inpatient Status and Location
Outpatient Appointments
What is VistA?
• Enterprise-wide source for Patient ID with business rules
– Advanced “Master Patient Index” (MPI)
• With consolidation capabilities for duplicate records
• Access of patient medical information stored in other facilities
• Powerful toolkits
– For database and client/server
– Multiple programming languages
•
•
•
•
•
•
8
Java
Visual Basic
Visual C++
Apache/CGI
Delphi
…and more to come
Why Use VistA?
• Modular
– Allows tailoring for the needs of any healthcare organization
• Improves quality of care
– Provides ALL patient data in one easy-to-access repository
– Facilitates communication among caregivers
• Allows data to be…
– Entered at the point of service
– Shared among different sites and/or providers
• Continuity of inpatient & outpatient records
• Reduces cost
–
–
–
–
9
Eliminates duplicate-orders
Reduces time spent searching for patient information
Eliminates redundant record-keeping
Simplifies data-collection for research and reporting
Why Use VistA?
• Reduces errors due to lack of data or poor handwriting
– Increases patient safety
• Provides enhanced data privacy and security protections.
• Automated Clinical Support…
– Clinical reminders (prevention, treatment,…)
– Alerts (notifications to clinician about real time issues)
– Scheduled and un-scheduled batch processing
• Tracks consults and orders from requests to results
10
Why Use VistA?
• Adheres to National HIT Standards
– “Consolidated Healthcare Informatics”
(CHI) Initiative
• DICOM
– Retrieval and transfer of images and
associated diagnostic information
• HL7
– Demographic information
– Units of measure
– Immunizations
– Clinical encounters
– Clinical Document Architecture for textbased reports
• LOINC
– EDI of clinical laboratory test orders
– EDI of clinical laboratory results
– Drug label section headers
• HIPAA
– ANSI X 12N
» Billing and Financial Standard Transactions
– NCPDP
» Retail Pharmacy Transactions
– CPT, HCPCS, ICD-9, NDC,…
» Standard Code-sets
11
• SNOMED CT
– Laboratory result contents
– Non-laboratory interventions and
procedures
– Anatomy
– Diagnosis and problems, and
– Nursing
• IEEE 1073
– Connect medical devices to information and
computer systems
– Allows monitoring of information remotely
• FDA
– Names and codes for ingredients
– Manufactured dosage forms
– Drug products
– Medication packages
• NLM RxNORM
– Describe clinical drugs
• VA National Drug File Reference
Terminology (NDF-RT)
– Drug classifications
Why Use VistA?
• Open Source / Public Domain
– Free (unless implemented on top of a commercial database)
• Significantly reduces Total Cost of Ownership (TCO)
• Fully-functional
• Fully-integrated
– HIS, RIS, LIS, PACS, …
• Enterprise-wide
• National-scale
• Multiple sources of support
– NOT proprietary…
– NOT tied limited support sources for an extended period of time…
12
Some VistA Components
• Clinical Functionality
–
–
–
–
–
–
–
–
–
–
–
–
Adverse Reaction Tracking
Clinical Reminders
Consults/Request Tracking
CPRS
Dentistry
Dietetics
Health Summary
Hepatitis C Extract
Home Based Primary Care
Imaging & Multimedia
Immunology Case Registry
Inpatient Medications
• Including Intravenous, Unit
Dose, etc.
– Intake and Output
13
– Laboratory
• Including EDI, Anatomic
Pathology, Blood Bank
–
–
–
–
–
–
–
–
Lexicon Utility
Medicine Clinical Services
Mental Health
Nursing
Oncology
Outpatient Pharmacy
Patient Care Encounter
Patient Information
• Admission, Discharge,
Transfer, Registration
Some VistA Components
•
Clinical Functionality (cont.)
– Pharmacy
• Automatic Replenishment /
Ward Stock
• Bar Code Medication
• Benefits Management (PBM)
• Controlled Substances
• Data Management (PDM)
• Drug Accountability /
Inventory
• Electronic Claims
• Mail Out Pharmacy
• National Drug File
• Outpatient Pharmacy
• Prescription Practices (PPP)
– Primary Care Management
14
–
–
–
–
–
–
–
Problem List
Prosthetics
Radiology/Nuclear Medicine
Remote Order Entry
Social Work
Spinal Cord Dysfunction
Surgery
• Including Risk Assessment
–
–
–
–
Visual Impairment
Vitals/Measurements
Women's Health
…
Some VistA Components
• Administrative
– Authorization/Subscription
Utility
– Automated Information
Collection System
– Automated Medical
Information Exchange
– Automated Safety Incident
Surveillance Tracking System
– Decision Support System
Extracts
– Duplicate Record Merge:
Patient Merge
– Engineering
– Equipment/Turn-In Request
– Event Capture
– Incident Reporting
15
– Library
– Master Patient Index
• Including Demographics
–
–
–
–
–
–
–
–
–
Network Health Exchange
Occurrence Screen
Police and Security
Record Tracking
Resident Assessment
Instrument
Scheduling
Survey Generator
Voluntary Timekeeping
…
Some VistA Components
• System
– FileMan
– Kernel
• Kernel Installation and
Distribution System (KIDS)
• Kernel ToolKit
–
–
–
–
–
MailMan
SQL Interface (SQLI)
Health Level Seven (HL7)
Patient Data Exchange (PDX)
National On-Line
Information Sharing
– Patch Module
16
– Remote Procedure Call
Broker
– List Manager
– Remote Data Views
– Victory Programming
Environment (VPE)
– …
CPRS
17
CPRS
• Computerized Patient Record System
– Patient centered view of record
– Dynamically updated client-server architecture
– Remote Procedure based
• Centralized updates of functionality
• Central data storage and validation
• Efficient code maintenance
– Graphic User Interface
• Transparent visualization of patient record
• Consistent user interface model
• Separation of View-Model-Controller
18
Clinical Reminders
19
Clinical Reminders
• Recommendations, not Practice Standards
– Disease Screening
– Immunization
– Patient Education recommendations
• Guidelines to assist in clinical practice
• Appropriate utilization based on
– clinical judgment
– patient's current status.
20
Consult/Request Tracking
21
Health Status
10/14/2004 15:50
******************
DOB: 07/13/1950
****************** CONFIDENTIAL Clinical Data (1y) SU MMARY
HSPATIENT,ONE
000 -31-9898
1A(1&2)
--------------------------
BDEM - Brief Demographics
--------------------------
Address: 394 N 700 E
Kaysville, UTAH 84037
Eligibility: SC LESS THAN 50%
Phone: 801 444 1264
Age: 54
Sex: MALE
Ethnicity: DECLINED TO ANSWER
Race: WHITE
PCMM Team:
PCMM Provider:
Analog Pager:
Inpat. Prov/Attn:
PRIMARY
HSProvider,One
801 -588-5555-3938
HSProvider,Two
Treating Facility
---------------------------ELY
-------------------------
Phone: 801 -588-2121
Phone: 801 -588-5030
Phone:
Type
Station
----------- ------CBOC
660GC
BADR - Brief Adv React/All
Last Seen
---------06/28/2000
-------------------------
Allergy/Reaction: DUST, STRAWBERRIES, CHOCOLATE, MILK, SHRIMP
------------------ ----------- PLL - All Problems ----------------------------2 Problems
ST PROBLEM
LAST MOD
PROVIDER
A Migraine (ICD 346.90), Onset 01/25/2000
03/28/2 004 HSProvider,One
A
Diabetes (ICD 250.00)
02/08/2004
------------------- RXOP - Outpatient Pharmacy (max 1 year)
HSProvider,One
-------------------
---------------------- ORC - Current Orders (max 1 year) ---------------------Item Ordered
REVIEW GERIATRICS Cons Bedside
Status Start Date
pend
Stop Date
05/14/2004 09:31
------------- PN - Progress Notes (max 50 occurrences or 1 year)
Title: CARDIOLOGY
AUTHOR: HSPROVIDER,ONE
IHD Lipid Profile:
Outside lipid profile in past year at another VA or non
Outside LDL >129 mg/dL
Date: March, 2004
Location: Outside Physician's Office
-------------
09/16/2004 21:07
-VA facility.
< THE ABOVE NOTE IS UNSIGNED >
- DRAFT COPY * DRAFT COPY * DRAFT COPY * DRAFT COPY * DRAFT COPY * DRAFT COPY
* END *
Press <RET> to continue, ^ to exit, or select component:
22
-
Federal VistA Adopters
• Veterans Administration (VistA)
• Department of Defense (CHCS)
• Indian Health Service (RPMS)
• U.S. Public Health Service
• NASA
23
US VistA Adopters
•
•
•
•
•
•
•
•
•
24
Washington D.C.
West Virginia
South Carolina
Oklahoma
Minnesota
Washington
City of Berkeley, CA
City and County of San Francisco
…and more to come
International VistA Adopters
• City of Bogotá, Colombia
• Germany
• Egypt
• Kenya
• American Samoa
• …and more to come
25
Recent VistA Embracers
• Panamá
• México
• Japan
• China
• Malaysia
• Jordan
• India
• …and still more to come
26
HardHats
• Virtual community of worldwide VistA users…
• Historical information about VistA
• Mailing-list
– Access to a worldwide community of VistA users
• Technical Information / Documentation
–
–
–
–
FileMan
M Utilities and Information
VA Utilities
Etc.
• VistA-distributions
– VistA under:
• InterSystems Caché
• Fidelity GT.M
27
Relationships Between US Government
and External Communities
Freedom of Information Act Releases
Vista Users
Distributions
U.S. Government
Veterans Affairs
Health & Human Services
WorldVistA Community
Collaborative
Development
Vista Programmers
Enhancements, Bug Fixes,
Etc.
28
M Technology
• MUMPS
– “Massachusetts General Hospital Utility Multi-Programming System”
– Also known as “M”
• Programming language, database management
system and related bindings and protocols
• Predominant world-wide in large integrated
hospital information systems.
– Also successful for systems in financial, travel,
shipping, and other industries.
29
M Technology
• Main features include:
–
–
–
–
–
–
–
–
30
Comprehensive procedural language
Support for object-style event-driven processing
Integrated hierarchical data management facility
Bullet-proof database security and transaction
processing
Multi-user, multi-tasking operating environment
Integrated inter-process communication
Support for distributed data and distributed processing
High-performance client-server networking capabilities
M Technology (cont)
• Available with “M” are:
–
–
–
–
–
–
–
31
Relational data manipulation tools
Industry-standard SQL
SQL-based connectivity
ODBC
Interfaces to windowing managers
ANSI X3.64 standard support for character devices
ANSI GKS standard support for graphics devices
What are Ontologies?
• Controlled Vocabularies and Taxonomies
– descriptions built from simpler components
– broad coverage of data values
• Formalized Rules and Facts
–
–
–
–
–
32
hierarchically arranged categories
first order logic formal semantics
specialization and generalization classification system
dynamic inference rules augmenting system
integrated search using rules and facts
How does VistA use Ontologies?
• Two Answers – both true
– VistA doesn’t use Explicit Formal Ontologies
• Long history of development (from 1980’s to present)
• Implements progressive ontological concepts as available
– VistA uses Ontologies throughout
•
•
•
•
•
•
•
33
Many ontological components used
Declarative (table driven) strategy is common
Hierarchical data structures with inheritance
Rule driven methodologies used
Expert system used for order checking
Controlled vocabularies for workload, diagnosis
Permission Security System is rule based
How does VistA use Ontologies?
• Implicit Ontology based on database definitions
and business rules
– VA FileMan Database has strong modeling capabilities
• Hierarchical, based on Codd and Date’s original theories
• Primary focus on clinical and medical information storage
• Consistency and Integrity maintenance
–
–
–
–
34
Database elements and user interface consistency
Procedural verification and action semantics
Declarative (table driven) implementation
Cross table and field constraint system
How does VistA use Ontologies
• Enterprise-wide consistent vocabularies
– Lexicon based on UMLS
– National Drug Formulary
– Ongoing Standardization Process
•
•
•
•
Drug Classes
Drug Ingredients
Allergy names
Packaging, Medical Routes
– ICD9 DRGs (545 entries)
– CPT codes /AMA procedures (~17,000 entries)
– Progress Note Titles and Hierarchy
35
How does VistA use Ontologies
Lexicon based on Unified Medical Lexicon System (UMLS)
– Table driven Vocabulary system
•
•
•
•
•
•
•
•
•
•
•
•
36
MAJOR CONCEPT MAP (~120,000 entries)
CONCEPT USAGE (~120,000 entries)
EXPRESSIONS (~153,000 entries)
EXPRESSION TYPE (7 entries)
EXPRESSION FORM (14 entries)
CODES (63,000 entries)
CODING SYSTEMS (33 entries)
SEMANTIC MAP (~174,000 entries)
SEMANTIC CLASS (15 entries)
SEMANTIC TYPE (133 entries)
SOURCE CATEGORY (487 entries)
etc.
How does VistA use Ontologies
Hierarchically Driven System
– User Menu based on Organization’s Functional Areas
• Role based security
• Menu Items based on day-to-day actions
– Graphical Interface (CPRS) Parameters
• Default values hierarchy
– individual, department, software subsystem, local hospital
– Values used for data entry and data presentation
• Rule driven capabilities
– Access to resources (orders, consults, progress notes)
– Students require co-signers
– Deleting, Modification, Viewing
37
How does VistA use Ontologies
Order Checks Expert System
38
•
ORDER CHECKS
•
ORDER CHECK COMPILER FUNCTIONS (56 entries)
•
ORDER CHECK DATA CONTEXT
•
ORDER CHECK DATA FIELD
•
ORDER CHECK DATA SOURCE
•
ORDER CHECK ELEMENT
•
ORDER CHECK NATIONAL TERM
•
ORDER CHECK PATIENT RULE EVENT (# entries dependent on site ~300,000)
•
ORDER CHECK RULE
(27 entries)
(7 entries)
(158 entries)
(15 entries)
(139 entries)
(14 entries)
(71 entries)
How does VistA use Ontologies
Order Checks Expert System
39
•
ORDER CHECKS
•
ORDER CHECK COMPILER FUNCTIONS (56 entries)
•
ORDER CHECK DATA CONTEXT
•
ORDER CHECK DATA FIELD
•
ORDER CHECK DATA SOURCE
•
ORDER CHECK ELEMENT
•
ORDER CHECK NATIONAL TERM
•
ORDER CHECK PATIENT RULE EVENT (# entries dependent on site ~300,000)
•
ORDER CHECK RULE
(27 entries)
(7 entries)
(158 entries)
(15 entries)
(139 entries)
(14 entries)
(71 entries)
Some Order Checks Currently Used
ALLERGY-CONTRAST MEDIA INTERACTION
ALLERGY-DRUG INTERACTION
AMINOGLYCOSIDE ORDERED
BIOCHEM ABNORMALITY FOR CONTRAST
MEDIA
CLOZAPINE APPROPRIATENESS
CRITICAL DRUG INTERACTION
CT & MRI PHYSICAL LIMITATIONS
DANGEROUS MEDS FOR PT > 64
DISPENSE DRUG NOT SELECTED
DUPLICATE DRUG CLASS ORDER
DUPLICATE DRUG ORDER
DUPLICATE OPIOID MEDICATIONS
DUPLICATE ORDER
ESTIMATED CREATININE CLEARANCE
40
GLUCOPHAGE-CONTRAST MEDIA
GLUCOPHAGE-LAB RESULTS
LAB ORDER FREQ RESTRICTIONS
MISSING LAB TESTS FOR ANGIOGRAM
PROCEDURE
NO ALLERGY ASSESSMENT
ORDER CHECKING NOT AVAILABLE
POLYPHARMACY
RECENT BARIUM STUDY
RECENT ORAL CHOLECYSTOGRAM
RENAL FUNCTIONS OVER AGE 65
SIGNIFICANT DRUG INTERACTION
Summary
VistA and Ontologies
• large Hospital Information System
• multiple sources of content suitable for inclusion
and formalization
• History of Innovative inclusion of technologies
• Consistent data model and management system
• Non-unified ontological approach
• Would benefit from Standard Upper Ontology
41
Thank You
Questions?
Chris Richardson & David Whitten
http://www.worldvista.org
June 16, 2005
42