Clinical Research Data Systems and Networks Supported by the BU-CTSI - - PowerPoint PPT Presentation

▶

Feb 02, 2024 371 likes •668 views

Clinical Research Data Systems and Networks Supported by the BU-CTSI William G. Adams, MD Professor of Pediatrics Director, BU-CTSI Clinical Research Informatics Boston University School of Medicine/Boston Medical Center badams@bu.edu Our

SLIDE 1

William G. Adams, MD

Professor of Pediatrics Director, BU-CTSI Clinical Research Informatics Boston University School of Medicine/Boston Medical Center badams@bu.edu

Clinical Research Data Systems and Networks Supported by the BU-CTSI

SLIDE 2

Our “EcoSystem”

BMC is largest safety net

provider in New England

Nearly all CHCs are FQHCs
EHR-based care
BMC since 1999
CHCs since 2003

SLIDE 3

A Vision

Data is open - privacy is protected
Researchers focus on questions more

than queries

Data and tools are standardized so

the same question can be asked at multiple sites

SLIDE 4

BMC Clinical Data Warehouse

Source data for BMC clinical systems
Dedicated research data analyst(s)
Includes CHC data and BMC data
Claims being added

SLIDE 5

BU-CTSI Clinical Data Resources and Networks

SLIDE 6

i2b2

“Informatics for Integrating Biology with

the Bedside”

Open-source software based on the MGH

Research Patient Data Repository (RPDR)

Collection of modules or “cells” constitute

the i2b2 “hive”

De-identified clinical data repository
Data linked to standardized vocabularies
Web-based query and analytic tools

SLIDE 7

SLIDE 8

Staging Area i2b2

Functions:

MPI linkage
Data cleaning
Standardization

(LOINC, CPT, RxNorm, ICD9,SNOMED CT) Data:

Demographic
Insurances
Services
Medications
Problems
Labs
Clinical Observations

EHR Data

Sites:

Boston Medical Center
Dorchester House MSC
Codman Square HC
Healthcare for the Homeless
Greater Roslindale MDC
Mattapan CHC
South End CHC
South Boston CHC
Uphams Corner CHC

Database

People (1.6+ mil)
Facts (2+ billion)
Concepts

Tools

Query Cell
HOME Cell

BMC-i2b2

BMC only
Web accessible
Aggregate data
No additional IRB

Health Disparities Repository

BMC + CHCs
RDP access (with SAS, STATA)
HOME Cell and data extracts
IRB approval required

BU-i2b2 Resources and Processes

SLIDE 9

Ontologies Facts Counts

SLIDE 10

I2b2 Data Model

SLIDE 11

i2b2 Compatible Data

Demographics
Problems/Diagnoses
Medications
Clinical Observations
Procedures
Laboratory Data
Genomic Data
Much more…

SLIDE 12

i2b2 Recent Additions

SLIDE 13

SLIDE 14

Sample ACT SHRINE Query

SLIDE 15

SLIDE 16

SLIDE 17

Sample Query: Diabetes and A1C Cohort

SLIDE 18

Sample Query: Narrowing Results Range

SLIDE 19

Sample Query: Counting Patients

SLIDE 20

Sample Query: Exploring the Cohort

SLIDE 21

Sample Query: Predicting Arrivals

SLIDE 22

SLIDE 23

Projects/activities

BMC Projects

BMC Cancer Registry Integration
BP Normalization and Care
VVV in BP, lipid, and Hgb1C
Sickle Cell QI (children and adults)
Community-based smoking cessation
Pneumonia rates in PCV vaccine era
HPV vaccination and morbidity
Algorithms for Personalized Decision

Making

Vital Village – geographic health effects
Asthma, Housing, Environment Models
Social Determinants and Child BH

I2b2 Networking

Cardiovascular Health Atlas
ePROS – psychotropic

medication use in kids

ePROS – on- off-label safety
ePROS – pediatric hypertension
Long-term outcomes following

bariatric surgery

Insurance switching
ACT Network
PMI/All of Us
Predicting suicide
OHDSI

SLIDE 24

SLIDE 25

I2b2 vs OMOP Data Models

OMOP i2b2

SLIDE 26

SLIDE 27

Opportunities/Challenges

High-touch vs self-service
Building capacity
Data literacy
Secure workspace(s)
Analytic support
Data governance
Asking compelling questions

SLIDE 28

ARCs

At least 5 investigators and trainees and a director
Focus on a research theme, explored with the aid
f different disciplines and technologies.
Selected ARCs funded for up to three years via

DOM+CTSI+IBRO

Key words: Inclusiveness; Innovation; Interdisciplinary

Time for a Health Data Science Affinity Research Collaborative (ARC) at BU/BMC?

Contact Bill Adams (Bill.Adams@bmc.org) or/and Katya Ravid (kravid@bu.edu) with ideas or for more information.

SLIDE 29

Investigator Expertise Roles

William Adams CRI, Epi, i2b2

Director, clinical and population health informatics lead, manages and promotes i2b2 and OMOP networks

Azer Bestavros CDS,ML-PA,EDS

Computational and data science lead

Marc Lenburg BI

Co-director, bioinformatics lead, CRC bioinformatics liaison

Rebecca Mishuris CRI, EHR, PH

EHR innovation research, ITS-liaison, Epic SME

Christopher Shanahan CRI,RD

CRITC lead, app and registry SME, addiction informatics SME

Ioannis Paschalidis CDS,ML-PA

Machine learning, prediction, School of Engineering liaison

Bindu Kalesan Epi, HSR

Translational Epidemiology and computational resources

Belinda Borelli MH, TBC

Mobile Health lead technology-based behavior change SME

Adam Gower BI

Bioinformatics analytic support, OpenSesame and GeneHive

Adam Labadorf BI

BU-HUB lead, bioinformatics analytic support

Michael Silverstein HSR, PH, ACO

Population Health and ACO Analytics lead, data governance

Martha Werler PHI, Epi, PH

Public Health Informatics, promotes Optum and other data

Nan Do CRI, EHR, HSR

VA-based informatics lead

Expertise Legend: ACO-Accountable Care Organization, BI-Bioinformatics, CDS-Computational and Data Sciences, CRI–Clinical Research Informatics, EHR-Electronic Health Record, Encryption and Data Security (EDS), HSR-Health Services Research, i2b2- Informatics for Integrating Biology and the Bedside, ML-PA-Machine Learning/Predictive Analytics, MH-Mobile Health, OMOP- Observational Medical Outcomes Project, PH-Population Health, RD-Registry Development, SME-subject matter expert, TBC- Technology-based Behavior Change