Twenty Years of IAIMS: The Columbia University/ New York Presbyterian Hospital Clinical Data Repository James J. Cimino Department of Biomedical Informatics.

Twenty Years of IAIMS:The Columbia University/

New York Presbyterian Hospital Clinical Data Repository

James J. Cimino

Department of Biomedical Informatics

Columbia University College of Physicians and Surgeons

IAIMS Consortium Annual Meeting

Boston, MA

April 10, 2005

• History and evolution (or creation)

• Where we are today

• What we learned

Overview

• History and evolution

History and Evolution (or Creation)

• 1983-1986: IAIMS Planning Grant– Rachel Anderson– Organizational

• 1986-1988: IAIMS Demonstration Grant– Paul Clayton and Rachel Anderson– Center for Medical Informatics– Vision

History and Evolution (or Creation)

• 1983-1986: IAIMS Planning Grant– Rachel Anderson

– Organizational

• 1986-1988: IAIMS Demonstration Grant– Paul Clayton

– Center for Medical Informatics

– Vision

– “$6M Demo”

• 1988-1993: IAIMS Implementation Grant– Funding from NLM, IBM, CU, Presbyterian Hospital– Network– Clinical data architecture

Clinical Data Architecture

• Central repository to collect data from myriad sources• Myriad users of data - some not yet imagined

New York Presbyterian HospitalClinical Information Systems Architecture

Clinical Database

Medical Entities Dictionary

Database Monitor

Medical Logic Modules

DatabaseInterface

Research

Administrative

Alerts & Reminders

Results Review

. . .. . .Radiology LaboratoryDischarge

Summaries

Reformatter Reformatter Reformatter

Clinical Data Architecture

• Central repository to collect data from myriad sources• Myriad users of data - some not yet imagined• Patient-oriented, not visit oriented, database• Relational, not hierarchical, model• Entity-attribute-value model

Entity-Attribute-Value Clinical Data Repository

Clinical Data Architecture

• Central repository to collect data from myriad sources• Myriad users of data - some not yet imagined• Patient-oriented, not visit oriented, database• Relational, not hierarchical, model• Entity-attribute-value model

• Coded data wherever possible• Unify terminology

Medical Entities Dictionary: A Central Terminology Repository

Communicating Terminology Changes

K#1

K#2 K#3

K#3 = 2.6

K#1 = 4.2K#1 = 3.3

K#2 = 3.2

K#1 = 3.0

K#1 = 4.2K#1 = 3.3

K#2 = 3.2

K#1 = 3.0

Solution: Hierarchical Integration

K#1

K#2

K

K#3

K#3 = 2.6

MED Structure

MedicalEntity

LaboratoryProcedure

CHEM-7PlasmaGlucose

LaboratorySpecimen

PlasmaSpecimen

Substance

Sampled

Part of

Has S

pecimen

Event

LaboratoryTest

DiagnosticProcedure

Substance MeasuredGlucose

Plasma

AnatomicSubstance

Substance

BioactiveSubstance

Chemical

Carbo-hydrate

Where We Are Today - Repository

• Patients: 2.6 million• Visits: >10 million since 1996 with

archives going back to 1979• Visit diagnoses, locations,

procedures, providers, insurance• Lab procedures: 16 million with 130

million results (to 1989)• Radiology procedures reports: 5.7

million• Pathology: 1.4 million• Cardiology procedures: 1.5 million • Resident signout notes:760,000• Operative Notes: 426,000• Clinical Notes: 400,000• Discharge Summaries: 420000

• Medication orders: >60 million• ObGyn Procedure Reports: 241,000• GI Procedure Reports: 101,000• Neurology Procedure Reports:

54,000 • Ideatel BP’s: 215,000• Ideatel Glucose: 650,000• Consult Events: 18000• HEENT Events:13000• Hospitalist Notes:30000• PFT: 25000• Provider profiles 11000• IDX 1.4 million• East Campus

Where We Are Today - MED• Domains:

– HP lab terms– Misys lab terms– Cerner lab terms– Misys Radiology– Digimedix drugs– Cerner Drugs– ICD9-based problem list terms– Other applications– Knowledge terms

• Size:– Concept-based (95,641)– Multiple hierarchy (141,306)– Synonyms (239,581)– Translations (141,717)– Semantic links (225,698)– Attributes (210,456)

Where We Are Today - Outputs

• 7000 Users• Clinical information systems

[LPRO]->(AE)->[ANTB]->(DS)->[PFUN]->(PO)->[ORGM]<-(PP)<-[OATT]

A procedure assesses the effect of an antibiotic which disrupts aphysiologic function which is a process of an organism which has an

attribute (sensitive/resistant).

Where We Are Today - Outputs

• 7000 Users• Clinical information systems• MedLEE

MedLEEHISTORY OF PRESENT ILLNESS: This 67 year old with a history of syncope in 1987 and 1989. She reported that she was evaluated both times and the work up was negative for any specific etiology. On the day of admission she reports having one episode of severe diarrhea and she was having increasing abdominal discomfort with flatulence and one episode of vomiting. When she returned to the bath room to move her bowels again she felt light headed and called for a family member. The family member reported that the patient was unconscious at that time and was placed in bed and recovered within 1-2 minutes. there was no history of any precipitating shortness of breath, chest pain or any seizure activity. At the time the patient was seen in the hospital she already felt fine.

Problems present: diarrheadiscomfort (abdomen ) vomitinglightheadedunconscious

Problems absent: pain (chest )seizure

Findings present: demo ( 67 year )

History: syncope

Where We Are Today - Outputs

• 7000 Users• Clinical information systems• MedLEE• Decision support systems

– Vigilens: TB, Freq Admit, Lab Vals

Where We Are Today - Outputs

• 7000 Users• Clinical information systems• MedLEE• Decision support systems

– Vigilens: TB, Freq Admit, Lab Vals– Infobuttons

Where We Are Today - Outputs

• 7000 Users• Clinical information systems• MedLEE• Decision support systems

– Vigilens: TB, Freq Admit, Lab Vals– Infobuttons

• Clinical data warehouse

2000 2001 2002 2003 2004

MI

MI+Beta

select patient_id , time = primary_time

from visit2004_diagnosis

where diagnosis_icd9_code like '410%'

and b.primary_time between '01/01/2000' and '01/01/2005'

and b.comp_code = 30366

Where We Are Today - Outputs

• 7000 Users• Clinical information systems• Decision support systems

– Vigilens: TB, Freq Admit, Lab Vals– Infobuttons

• Clinical data warehouse• Other clinical systems

– Infection Control– CPOE– Marconi– IDEATel– Data Mining– Bioinformatics

Lessons Learned

• The repository architecture paid off

• Model the data, not the applications

• Write once, read many times

• Pay attention to your terminology

• You will reuse data

• You can’t predict how you will reuse it

Acknowledgements

• People:– Tom Morris– Henrik Bendixen– Rachel Anderson– Paul Clayton– Steve Johnson– George Hripcsak– Bob Sideli– Somitra (Sen) Sengupta

• New York Presbyterian Hospital• Columbia University• National Library of Medicine• IBM