health information exchanges and big data: challenges … · hadi kharrazi, mhi, md, phd johns...

Hadi Kharrazi, MHI, MD, PhD

Johns Hopkins [email protected]

Health Information Exchanges and Big Data: Challenges and Opportunities

Big Data and Analytics in HealthcarePhiladelphia 2013

© Hadi Kharrazi @ JHSPH-HPM

Big Data and HIEs

2

Introduction

Big Data and Healthcare

HIE History, Architecture and Services

HIE Examples (Indiana HIE, CRISP)

HIE Big Data Exploration (translating CPGs into population health data)

HIE and Population Health IT Framework

JHU Center for Population Health IT (CPHIT)

The presentation in a nutshell


Big Data and HIEs

3

• Hadi Kharrazi, MHI, MD, PhD

o Assistant Professor, Johns Hopkins School of Public HealthDepartment of Health Policy and Management

o Joint Appointment, Johns Hopkins School of MedicineDivision of Health Sciences Informatics

o Assistant DirectorJohns Hopkins Center for Population Health IT (CPHIT)

o E: [email protected] L: http://www.linkedin.com/in/kharrazi/o W: http://jhsph.edu/cphit

o Research interests: Population DSS, HIE & ACO Analytics

Introduction


Big Data and HIEs

4

Statistics:o Data size: 2012 = 500 petabytes 2020 = 25,000 petabyteso Cost: 2009 = 17.6% GDP ($2.9t) 2025 = 25.0% GDPo Big Data Saving: ~$300b/yr

Definition:o Big Data is a collection of data sets so large and complex that it becomes

difficult to process using on-hand database management tools or traditional data processing applications.

History in Healthcareo Last time: We pushed the data boundary in the Human Genome Project (1990)o This time: Massive roll out of EHRs (Meaningful-Use) and Integration of

various source of data (genomic, mobile, patient-generated, exchanges)

Big Data and Healthcare


Big Data and HIEs

5

Big Data Specs – data driven 4Vs:o Volume quantity (size)o Variety type (structure, standardized, ontology)o Veracity quality (meaning, completeness, accuracy)o Velocity time (real-time, timeliness)

Big Data and Healthcare (cont.)

a b c d e …

?

Volume

Veracity

Variety

Velocity


Big Data and HIEs

6

Molecular Research Health Research

Biomedical informatics as a basic science

Basic Research

Applied Research

Biomedical informatics methods, techniques, and theories

Bioinformatics

ImagingInformatics

ClinicalInformatics

Public HealthInformatics

Consumer HealthInformatics

PopulationHIT



Big Data and HIEs

7

Big Data Specs – clinical driven 5Ms:o Measure size, type, quality (QMs, digitization, …meaningful use)o Mapping integration, interoperability (information exchanges, EDWs)o Methods analytics (exploration, visualization, predictive, …replacing RCTs)o Meanings knowledge (EBM, CPG, …meaningful use)o Matching outcomes (triple aims, ACA)


Mea

sure

Map

ping

Met

hods

Mea

ning

Mat

chin

g

Copyright H.Kharrazi

EHRs

Genomics

Bio Signal

Consumer

Public

MandatesCQMs

Devices

EDWs

ACOs

HIEs

NwHIN

MandatesBusinessStandards

Expl/Visu

Statistical

Data Mining

Machine Learning

AnalyticsModelingSimulation

ApplicationKnowledge Outcome

CPG

CDSS

EBM

Theories

Triple Aim

ACA

HITECH

MU1,2,3

Otherpolicies

? ?? ??? ????? ??


Big Data and HIEs

8

Johns Hopkins


JH Hospitals

Copyright H.Kharrazi

JH CommunityPhysicians

JH Health Care

JCHIP / JCASE (CMMI)

Measure

Mapping

Methods

Meaning

Matching

JH A

CO

Epic

-

-

??

???

??


Big Data and HIEs

9

HIE History

Copyright HiMSS


Big Data and HIEs

10

HIE History (cont.) > ONC

Copyright HiMSSONC relationship with DHHS


Big Data and HIEs

11

HIE (verb): The electronic movement of health-related information among disparate organizations according to nationally recognized standards in an authorized and secure manner.

HIO (noun): An organization that oversees and governs the exchange activities of health-related information among independent stakeholders and disparate organizations according to nationally recognized standards in an authorized and secure manner.

An HIO can be described by many acronyms, including:

o State Level Health Information Exchange (SLHIE)

o Regional Health Information Exchange (RHIO)

o Regional Health Information Network (RHIN)

o Health Information Exchange Networks (HIE[N])

o Others: Integrated Delivery Systems (IDNs); Physician practices HIEs; Payer-led HIEs; and, Disease-specific HIEs.

HIE History (cont.) > Health Information Organization (HIO)

Copyright HiMSS


Big Data and HIEs

12

HIE Architecture > Centralized

Copyright Regenstrief InstituteCentralized HIE

Data Repository

EHRs

Claims

PACS LABs

Registries

PHRs


Big Data and HIEs

13

HIE Architecture (cont.) > Federated

Copyright Regenstrief InstituteFederated Inconsistent HIE

EHRs

Claims

PACS LABs

Registries

PHRs


Big Data and HIEs

14

HIE Architecture (cont.) > Federated

Copyright Regenstrief InstituteFederated Consistent HIE

MPI

EHRs

Claims

PACS LABs

Registries

PHRs


Big Data and HIEs

15

HIE Architecture (cont.) > Hybrid

Copyright Regenstrief InstituteHybrid HIE

MPI

EHRs

Claims

PACS LABs

Registries

Data Repository

EHRs

LABs

Claims

PHRs


Big Data and HIEs

16

Data Standardization

Machine

HIE Architecture (cont.) > Switch

Copyright Regenstrief InstituteSwitch HIE

EHRs

Claims

PACS LABs

Registries

PHRs


Big Data and HIEs

17

HIE Architecture (cont.) > Patient Centric

Copyright Regenstrief InstitutePatient Centric HIE (e.g., PHR controlled)

EHRs

Claims Registries

Other…

LABsPACS


Big Data and HIEs

18

HIE Architecture (cont.)

Copyright Regenstrief InstituteHealth Information Organizations / Regional HIOs


Big Data and HIEs

19

HIE Architecture (cont.) > NwHIN

Copyright Regenstrief InstituteConnecting RHIOs / NwHIN


Big Data and HIEs

20

HIE Architecture (cont.) > NwHIN

Copyright HiMSSNationwide Health Information Network (NwHIN)

NHII (2004) NHIN (2010) NwHIN (2011)


Big Data and HIEs

21

Presentation Services: login, patient look-up, request patient records, view data

Business Application Services: e-Prescribing, EMR, lab, radiology, eligibility checking, problem list/visit history

Data Management Services: data persistence/access, value/code sets, key manage.

Data Storage Services: message logs, XML Schemas, Provider/User Directory

Integration Services: message translation/transport, HL7 mapping, EMR adapter

System Management Services: system config, audit/logging, exception handling

HIE Services > Core Services

Copyright HiMSS

Measure

Mapping

Methods

Meaning

Matching

-

-

?

?

?


Big Data and HIEs

22

Hospitals:

o Clinical messaging

o Medication reconciliation

o Shared EHR

o Eligibility checking

Physicians:

o Result reporting

o Secure document sharing

o Shared EHR

o Clinical decision support

o Eligibility checking

Laboratory:

o Clinical messaging

o Orders

HIE Services > Data Services by Constituency

Copyright HiMSS

Public Health:

o Needs assessment

o Biosurveilance

o Reportable conditions

o Results delivery

Consumers:

o Personal Health Records

Researchers:

o De-identified longitudinal clinical data

Payers:

o Quality measure

o Claims adjustment

o Secure document transfer

Measure

Mapping

Methods

Meaning

Matching

?

??

??

?

-


Big Data and HIEs

23

Next Generation Analytics

o Data warehouse, data analytics and business intelligence

o Quality reporting support

o Performance management

o Fraud and abuse identification and prevention

o Care gap identification

o Care and disease management

o Public health monitoring and analysis

o Population monitoring and predictive profiling

HIE Services > Emerging Services

Copyright HiMSS

Measure

Mapping

Methods

Meaning

Matching

-

-

???

???

??


Big Data and HIEs

24

The Indiana HIE (IHIE) includes (as of mid-2011):

• Federated Consistent Databases

• 22 hospital systems ~70 hospitals

• 5 large medical groups and clinics & 5 payors

• Several free-standing labs and imaging centers

• State and local public health agencies

• 10.75 million unique patients

• 20 million registration events

• 3 billion coded results

• 38 million dictated reports

• 9 million radiology reports

• 12 million drug orders

• 577,000 EKG tracings

• 120 million radiology images

Copyright Regenstrief Institute

HIE Examples > IHIE


Big Data and HIEs

25

Focus Areas:

o Query Portal Growth

o Direct Secure Messaging

o Encounter Notification System (ENS)

o Encounter Reporting System (ERS)

o Health Benefits Exchange integration

HIE Examples (cont.) > CRISP (Chesapeake Regional Info. Sys. for our Patients)

Copyright CRISP

Progress Metric Result

Organizations Live

Hospitals (Total 48) 48

Hospital Clinical Data Feeds (Total 143 - Lab, Radiology, Clinical Docs)

86

National Labs 2

Radiology Centers (Non-Hospital) 5

Identities and Queries

Master Patient Index (MPI) Identities ~4M

Opt-Outs ~1500

Queries (Past 30 Days) ~3500

Data Feeds Available

Lab Results ~16M

Radiology Reports ~5M

Clinical Documents ~2M


Big Data and HIEs

26

AHRQ’s NGC (guideline.gov)

*Blood Pressure, Weight/Body Mass Index (BMI) - Every visit. For adults: Blood pressure target goal <130/80 mm Hg; BMI (body mass index) <25 kg/m2.

For children: Blood pressure target goal <90th percentile adjusted for age, height, and gender; BMI-for-age <85th percentile.

Foot Exam (for adults) - Thorough visual inspection every diabetes care visit; pedal pulses, neurological exam yearly.

Dilated Eye Exam (by trained expert) - Type 1: Five years post diagnosis, then every year.

Depression - Probe for emotional/physical factors linked to depression yearly; treat aggressively with counseling, medication, and/or referral.

Arden Syntax:

logic:if (creatinine is not number) or (calcium is

not number) or(phosphate is not number) then

conclude false;endif;product := calcium * phosphate;

-----------------------------------------------------CARE Language:

BEGIN BLOCK CHOLESTEROL

DEFINE "MALE_PT" {VALUE} AS "SEX" IS = 'M'DEFINE "CHOLESTEROL LAST" AS LAST "CHOL TESTS" [BEFORE "DATE"] EXIST IF "CHOLESTEROL LAST" WAS AFTER 5 YEARS AGOTHEN EXIT CHOLESTEROLELSE CONTINUE

Computerized CPGs (Know.Rep.Lang.)

mysql_connect(‘localhost’, ‘root’, ‘pass’, ‘ehr');

mysql_query(‘SELECT patient_id FROM patient WHERE patient.Fname = ‘peter’ AND patient.Lname = ‘Johnson’’);

$results = mysql_fetch();

$patient_id = $results[‘patient_id’];

mysql_query (‘SELECT * FROM lab WHERE patient_id = $patient_id’ and lab_term = ‘HBA1c’);

$results = mysql_fetch();

$lab_result = $results[‘lab_term’];

IF ($lab_result > 9){ECHO “Patient has a high HBA1c level”;

}

Computerized DSS

Patient has diabetes and has not had an eye exam in two years. Based on guideline xxx you may want to consider asking for an eye consultation.

Popup alert

HIE Big Data Exploration > Translating CPGs into Population Health Data

Clinical Practice Guidelines

Computerized CPG(CARE, Arden syntax)

Manual Programming for each CPG

Relational DB

One patient at the time


Big Data and HIEs

27

HIE Big Data Exploration (cont.)

Clinical Practice Guidelines

Computerized CPG(CARE, Arden syntax)

Manual Programming for each CPG

Relational DB 1

One patient at the time All patient at the time

Automated for all CPGs

CDSS SQL 1

CDSSSQL 2

CDSSSQL…

CDSSSQL n

Relational DB 2

Relational DB…

Relational DB n

NQF QM eMeasures

Population Health Informatics


Big Data and HIEs

28


Generating Renewable CDSS

CPG NQF

CAREArden

XMLCDSSSQL 1

CPG

Con

sort

ium

Syntactic Mapping

Semantic Integration

Relational DB 1

CDSSSQL 2

CDSSSQL…

CDSSSQL n

Relational DB 2

Relational DB…

Relational DB n

Rene

wab

le

Lexical Integration

XSLT trans.

Pragmatic Integration


Big Data and HIEs

29


Language Specific Language Free


Big Data and HIEs

30


Rules Population One patient*

CD4_1.sql 59 0.18CHOLESTEROL_1.sql 180 0.11CHOLESTEROL_2.sql 1 0.12DIABETES_1.sql 541 0.56DIABETES_2.sql 1561 0.47DIABETES_3.sql 182 0.40DIABETES_4.sql 121 0.40DIABETES_5.sql 1620 1.51

…. … …

TCL_2.sql 67 0.12TCL_3.sql 69 0.18TCL_4.sql 61 0.17TETANUS_SHOT_1.sql 1 0.07WHOLE_1.sql 127 0.16WHOLE_2.sql 129 0.13

Average (sec) 1249 0.73

Runtime per second

* each row shows average of 20 individual runs


Big Data and HIEs

31


Some rules are never hit for institute 1 (not shown in this diagram)

Applying all applicable CARE rules for Institute-1

0

10000

20000

30000

40000

50000

60000

70000

OBFO

RM_1.sql

OBFO

RM_3.sql

OBFO

RM_2.sql

OBFO

RM_5.sql

MEDICIN

E_2.sql

PREVENT_PAP_1.sql

PREVENT_STO

OL_2.sql

CHOLESTERO

L_2.sql

CHOLESTERO

L_1.sql

PREVENT_M

AM_2.sql

PNVX2_1.sql

FLU_SH

OT_4.sql

HEPVAX_B_1.sql

FLU_SH

OT_3.sql

PEDS_10.sql

PEDS_11.sql

PEDS_1.sql

PEDS_7.sql

PEDS_8.sql

PEDS_9.sql

TETANUS_SH

OT_1.sql

LIPID_MAN

AGEM

ENT_1.sql

TCL_2.sql

FLU_SH

OT_1.sql

PREVENT_M

AM_3.sql

K_2.sql

DIABETES_1.sql

PREVENT_STO

OL_4.sql

WHO

LE_2.sql

TCL_1.sql

DIABETES_5.sql

FLU_SH

OT_5.sql

K_1.sql

PREVENT_M

AM_1.sql

OBFO

RM_6.sql

IDC_REM_1.sql

TCL_4.sql

WHO

LE_1.sql

CD4_1.sql

PREVENT_STO

OL_1.sql

TCL_3.sql

HIV_PT_1.sql

MEDICIN

E_1.sql

Institute 1: All CDS Hits (regardless of missed proportion)

applicable CARE rules


Big Data and HIEs

32


Institute versus Rules (Raw Numbers) (zero rules / zero institutes are removed)

institutes

rules


Big Data and HIEs

33

Manual process of deciphering the rules (hit-means-miss OR hit-means-hit-only)The lower the better (less missed)

Missed applicable CARE rules for institute-1(CDSS-generated QM fingerprint)

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

HIV_PT_1.sql

MEDICIN

E_1.sql

PREVENT_M

AM_1.sql

TCL_3.sql

WHOLE_1.sql

DIABETES_1.sql

PEDS_8.sql

FLU_SH

OT_5.sql

TCL_2.sql

TETANUS_SHO

T_1.sql

MEDICIN

E_2.sql

TCL_4.sql

FLU_SH

OT_4.sql

K_1.sql

OBFO

RM_5.sql

PREVENT_STO

OL_2.sql

CHOLESTERO

L_2.sql

FLU_SH

OT_3.sql

PREVENT_STO

OL_1.sql

PREVENT_PAP_1.sql

DIABETES_5.sql

OBFO

RM_1.sql

PEDS_10.sql

PNVX2_1.sql

OBFO

RM_6.sql

OBFO

RM_2.sql

CHOLESTERO

L_1.sql

WHOLE_2.sql

PREVENT_STO

OL_4.sql

LIPID_MAN

AGEMEN

T_1.sql

PEDS_11.sql

IDC_REM_1.sql

PEDS_9.sql

FLU_SH

OT_1.sql

OBFO

RM_3.sql

K_2.sql

CD4_1.sql

TCL_1.sql

PEDS_1.sql

PEDS_7.sql

PREVENT_M

AM_3.sql

PREVENT_M

AM_2.sql

HEPVAX_B_1.sql

Institute 1: Percentage of Hit / Miss Rules

Percent

%13.49Avg


~12.3%



Big Data and HIEs

34

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

CD4_1.sql

FLU_SH

OT_1.sql

HIV_PT_1.sql

IDC_REM_1.sql

OBFO

RM_6.sql

PEDS_7.sql

PEDS_9.sql

TCL_4.sql

WHOLE_2.sql

PREVENT_PAP_1.sql

FLU_SH

OT_4.sql

PEDS_1.sql

PEDS_8.sql

OBFO

RM_2.sql

TETANUS_SHO

T_1.sql

PEDS_11.sql

OBFO

RM_3.sql

MEDICIN

E_2.sql

CHOLESTERO

L_1.sql

CHOLESTERO

L_2.sql

FLU_SH

OT_3.sql

K_2.sql

K_1.sql

PREVENT_M

AM_3.sql

OBFO

RM_1.sql

PEDS_10.sql

PNVX2_1.sql

OBFO

RM_5.sql

TCL_2.sql

LIPID_MAN

AGEM

ENT_1.sql

HEPVAX_B_1.sql

PREVENT_STO

OL_4.sql

PREVENT_STO

OL_2.sql

PREVENT_M

AM_2.sql

Institute 2: Inverse Percentage of Hit / Miss Rules

Percent

%3.47Avg

Manual process of deciphering the rules (hit-means-miss OR hit-means-hit-only)The lower the better (less missed)

Missed applicable CARE rules for institute-2(CDSS-generated QM fingerprint)


~5.5%



Big Data and HIEs

35

The effect size of the missed rules depends on the population size.

Missed rate by institute

institutes

rules

RulesMissed %



Big Data and HIEs

36

Population Health Info.

HIE and Population HIT > Population Health Informatics

Population Health Data-warehouse

HIE Public Health Informatics

A

B

C

Publ

ic H

ealth

Clin

ical

Inf

orm

atic

s

Personal Health

Records

Population Health Analytics

Insurance Data

Admin data repositoriesRx Data

CI

to P

ubH

I

CI

to P

ubH

I

PubH

Ito

CI

PubH

Ito

CI

EHRs

1…n


Big Data and HIEs

37

MPI

HIE and Population HIT > CDS Continuum (cont.)

Population-based CDSS across databases

Population-based HIE

EHRs

Claims

PACS LABs

Registries

PHRs


Big Data and HIEs

38

The Johns Hopkins Center for Population Health Information Technology

(CPHIT, or “see-fit”)

The mission of this innovative, multi-disciplinary R&D center is to improve the health and well-being of populations by advancing the state-of-the-art of Health IT across public and private health organizations and systems.

CPHIT focuses on the application of electronic health records (EHRs), mobile health (m-health) and other e-health and HIT tools targeted at communities and populations.

Director: Dr. J. Weiner

www.jhsph.edu/cphit

CPHIT

Copyright Dr. J. Weiner


Big Data and HIEs

39

CPHIT (cont.)


CPHIT

JH Health System

External PH/IDS Orgs.

JHU Academic Departments

and other R&D centers

JH Healthcare Solutions,

LLC

BusinessPartners

IndustryFoundations Government

CPHIT Organizational Linkages


Big Data and HIEs

40

CPHIT’s R&D Priorities

Development and testing of measures created from EHRs and other HIT systems to quantify quality, outcome and need within target populations and communities.

Use and advancement of computing methodologies – including natural language processing (NLP) and pattern recognition tools – to improve the application of non-structured EHR data for population-based interventions.

Initiation of effective approaches for linking provider-centric EHR systems with consumer-centric internet and mobile-based e-health applications.

Development of EHR-based tools and decision support applications to identify and help manage high risk populations for preventive and/or chronic care and to coordinate care within IDS/ACO.

Strategic approaches for creating an interoperable community of EHR networks and integrating them with the current functions of public health agencies (e.g., surveillance and vital records).

Creation of legal/ethical and policy frameworks to support the development of secondary use of EHRs for public health programs and research.

CPHIT (cont.)



Big Data and HIEs

41

Developing and Testing "e-ACGs": Using EHR / HIT Data to Enhance Johns Hopkins ACGs' Ability to Measure Risk and Health Status

Real-time, Cross-Provider High Readmission Risk Detection and Notification System

Evaluate the economic impact of the Shared Patient Virtual Health Record (SPVHR) in Inpatient Settings

Prescription Drug Monitoring Program (PDMP) Evaluation

Developing and evaluating an Inter-Provider Health Screening and Quality Measure System (IPHSQMS)

Identifying high risk pregnancies by linking claims and clinical data using natural language processing tools

www.jhsph.edu/cphit

CPHIT (cont.) > Research Pipeline


Big Data and HIEs

42

Thank you

Q & A

www.jhsph.edu/cphit