1

Evolution of a Clinical Research Evolution of a Clinical Research Informatics Group within a Informatics Group within a

ServiceService--oriented Clinical Trials oriented Clinical Trials Data Management OrganizationData Management Organization

B. McCourt, D. B. McCourt, D. FastesonFasteson--Harris, Harris, S. S. ChakrabortyChakraborty, C. , C. BovaBova HillHill

AMIA CRI SummitAMIA CRI SummitSan Francisco, March 2010San Francisco, March 2010

TopicsTopics

ContextContext

ChallengesChallenges

Organizational Design SolutionOrganizational Design Solution

2 Year Experience2 Year Experience

Now and next stepsNow and next steps

2

DCRI ContextDCRI Context

What is DCRI?What is DCRI?

DCRIDCRI is the largest academic is the largest academic clinical research organization clinical research organization (ARO) in the world(ARO) in the world

A A globalglobal coordinating center for coordinating center for multimulti--center clinical trials that center clinical trials that integrates the medical expertise of integrates the medical expertise of Duke University Medical Center Duke University Medical Center with the operational capabilities of a with the operational capabilities of a fullfull--service CROservice CRO

3

DCRI FactsDCRI Facts

Founded in 1969 with the development of the Duke Founded in 1969 with the development of the Duke Databank for Cardiovascular Diseases Databank for Cardiovascular Diseases

21 years of experience in coordinating multi21 years of experience in coordinating multi--center trials center trials in over 20 therapeutic areasin over 20 therapeutic areas

900+ staff and 120 clinical/statistical faculty900+ staff and 120 clinical/statistical faculty

FullFull--Service CapabilitiesService Capabilities

4,600 manuscripts in peer4,600 manuscripts in peer--reviewed journalsreviewed journals

More than 420 projects completed in 64 countries More than 420 projects completed in 64 countries enrolling more than 579,900 patientsenrolling more than 579,900 patients

DCRI DCRI –– Trials Experience by Phase and SizeTrials Experience by Phase and Size

4

CDM ChallengesCDM Challenges

History (1997 History (1997 –– 2007)2007)

History & Organizational DesignHistory & Organizational Design

Causal IssuesCausal Issues•• Capacity for changeCapacity for change•• Industry trendsIndustry trends

Environmental IssuesEnvironmental Issues•• Organizational changesOrganizational changes•• Qualitative studiesQualitative studies

5

Existing Org Structure Existing Org Structure (1997(1997--2007)2007)

Clinical Data ManagersClinical Data Managers

Data Management TeamsData Management Teams

Quality ControlQuality Control

Case Report Form DesignCase Report Form Design

Clinical ProgrammingClinical Programming

Duke FollowDuke Follow--upup

Medical CodingMedical Coding

Clinical Data Integration

Highly nested team unitsHighly nested team units

10 Year Headcount Trend10 Year Headcount Trend

6

Adopted newnetwork

Large paper trials locked

10 Year Headcount Trend10 Year Headcount Trend

Industry Trends Industry Trends (2007)(2007)

Increasing number of projects with Increasing number of projects with increasing complexity and decreasing data increasing complexity and decreasing data processing processing

Increased demand for EDC (vs. paper) and Increased demand for EDC (vs. paper) and increased demand for CDI consulting with increased demand for CDI consulting with sponsor EDC tools & implementations sponsor EDC tools & implementations

Increasing infrastructure initiatives Increasing infrastructure initiatives supporting DCRI, DTMI, NIH & Industry to supporting DCRI, DTMI, NIH & Industry to develop and adopt innovative operational develop and adopt innovative operational methods. methods.

7

Industry Trends Industry Trends (cont)(cont)

Increased reliance on standards and Increased reliance on standards and technology to meet increasing variety of technology to meet increasing variety of CDM requirementsCDM requirements

Emerging industry recognition of increasing Emerging industry recognition of increasing CDM scope and trendsCDM scope and trends

Environmental StudiesEnvironmental Studies

DCRI wide telephone surveyDCRI wide telephone survey

Departmental web surveyDepartmental web survey

Horizontal focus groupsHorizontal focus groups

8

What did the studies tell us?What did the studies tell us?

Opportunities for improvement:Opportunities for improvement:•• Inconsistent management decisions Inconsistent management decisions •• Career progression Career progression •• More effective communications More effective communications •• TrustTrust•• CollaborationCollaboration

Mastery of administrative, technical, training Mastery of administrative, technical, training and project management was difficult to and project management was difficult to achieve achieve

Organizational Design SolutionOrganizational Design Solution

9

Resulting Org Structure Resulting Org Structure (Dec 2007)(Dec 2007)

Clinical Data ManagersClinical Data Managers

Data Management TeamsData Management Teams

Quality ControlQuality Control

Case Report Form DesignCase Report Form Design

Medical CodingMedical Coding

Research InformaticsResearch Informatics

Clinical ProgrammingClinical Programming

Duke FollowDuke Follow--upup

Clinical Data Integration

Clinical Data ManagementDebra Fasteson-Harris

Associate Director

Clinical Research InformaticsBrian McCourt

Associate Director

Impact of Impact of matrixedmatrixed CDM teamsCDM teams

Administrative Reporting RelationshipsAdministrative Reporting Relationships•• 90 of 138 Employees switching managers90 of 138 Employees switching managers

Project Teams IntactProject Teams Intact•• 88 projects total88 projects total•• 69 have no changes to CDM Lead69 have no changes to CDM Lead•• 12 the new lead has been involved already 12 the new lead has been involved already

and lead role being formalizedand lead role being formalized•• 4 leads will change after resources available4 leads will change after resources available•• 2 new projects not yet assigned2 new projects not yet assigned•• 1 project transition is being scheduled1 project transition is being scheduled

10

DCRI Research InformaticsDCRI Research Informatics

Support research data integration projects Support research data integration projects with complex research requirements. with complex research requirements.

Evaluate and Evaluate and operationalizeoperationalize new ideas for new ideas for data management tools and methods (‘data data management tools and methods (‘data management pipeline’)management pipeline’)

Develop and implement clinical data Develop and implement clinical data standardsstandards

10 Year Headcount Trend10 Year Headcount Trend

11

Emergence of CRIEmergence of CRI

Papers on issuesPapers on issues

Clinical Research Informatics contextClinical Research Informatics contextDuke Translational Medicine Institute (DTMI)

DTRI DCRU DCRI DCCRDNRI GHI

12

2 Year Experience2 Year Experience

class Ortel

SAMPLES_4488

«column» SPECMCD: VARCHAR2(40) FK TRANSMID: NUMBER(6) SPECMNUM: VARCHAR2(20) SPECSEQ: VARCHAR2(10) SUBJID: VARCHAR2(20) VISIT: VARCHAR2(40) VISITNUM: VARCHAR2(20) = NULL VISITTYP: VARCHAR2(11) = S LBDTM: VARCHAR2(25) = NULL STUDNAM: VARCHAR2(200) STUDYID: VARCHAR2(20) = 4488 LBSPEC: VARCHAR2(40) = PLA S PROCNAM: VARCHAR2(100) SHPPLBDT: DATE SHPTO: VARCHAR2(50)

«FK»+ FK_SAMPLES_4488_TRANSMISSION(NUMBER)

«unique»+ UQ_4488_SAMPLES_SPECMCD(VARCHAR2)

ELISA_AV G_DATA

«column»*PK AVGDATID: NUMBER(10) SPECMCD: VARCHAR2(40)*FK PLATEID: NUMBER(6)*FK TESTID: NUMBER(6) RPTRESC: VARCHAR2(2048) = NULL RPTRESN: FLOAT CALSD: FLOAT CALCV: FLOAT RPTRFLG: VARCHAR2(10)

«FK»+ FK_ELISA_AVG_DATA_ELISA_PLATE(NUMBER)+ FK_ELISA_AVG_DATA_TEST(NUMBER)

«PK»+ PK_ELISA_AVG_DATA(NUMBER)

CLOTTING_ DATA

«column» CLDATID: NUMBER(6)*FK TESTID: NUMBER(6)*FK TRANSMID: NUMBER(6) SPECMCD: VARCHAR2(40) RPTRESC: VARCHAR2(2048) RPTRESN: FLOAT RPTU: VARCHAR2(20) TSTDTM: DATE TSTTYP: VARCHAR2(22) = S TSTDESC: VARCHAR2(40) TSTSTAT: VARCHAR2(13) = D SOFTWARE: VARCHAR2(40) = ACL TOP 2.8.7 INSTRMT: VARCHAR2(40) = ACL TOP INSTMSER: VARCHAR2(40) = 04090184 BATTRNAM: VARCHAR2(40) = Functional Assay BATTRID: VARCHAR2(20) = AIM1 RPTRTYP: VARCHAR2(12) = N RPTRSTAT: VARCHAR2(11) = F

«FK»+ FK_CLOTTING_DATA_TEST(NUMBER)+ FK_CLOTTING_DATA_TRANSMISSION(NUMBER)

COMMENT_4488

«column» SPECMCD: VARCHAR2(40) FK TRANSMID: NUMBER(6) SPECCOM: VARCHAR2(2048) SPECCND: VARCHAR2(2048) PLATENAM: VARCHAR2(40) TSTCOM: VARCHAR2(2048) TECHNAME: VARCHAR2(40) COMMDATE: DATE DILUTION: NUMBER(6) PLBNUM: VARCHAR2(20) = ORTEL STUDYID: VARCHAR2(20) = 4488 BATTRID: VARCHAR2(20)

«FK»+ FK_COMMENT_4488_TRANSMISSION(NUMBER)

ELISA_ PLATE

«column»*PK PLATEID: NUMBER(6)* TRANSMID: NUMBER(6) PLATENAM: VARCHAR2(40) TSTDTM: DATE TSTTYP: VARCHAR2(22) = S TSTDESC: VARCHAR2(40) TSTSTAT: VARCHAR2(13) = D RPTRTYP: VARCHAR2(12) = N BATTRNAM: VARCHAR2(40) = El isa Quantific... BATTRID: VARCHAR2(20) = AIM2 SOFTWARE: VARCHAR2(4 0) INSTRMT: VARCHAR2(40) INSTMSER: VARCHAR2(40) RPTRSTAT: VARCHAR2(11) = F FILENAME: VARCHAR2(40) FILCRDTM: DATE STUDYID: VARCHAR2(20) = 4488

«PK»+ PK_ELISA_PLATE(NUMBER)

TRANSMISSION

«column»*PK TRANSMID: NUMBER(6)*FK LABID: NUMBER(5) VERSION: VARCHAR2(7) = V1.0.01 TRMSRNUM: VARCHAR2(20) = ORTEL TRMSRNAM: VARCHAR2(40) = Duke Hemostasis... TRMTYP: VARCHAR2(11) = I FILENAME: VARCHAR2(40) LACTDTM: VARCHAR2(25) RECEXTYP: VARCHAR2(25) = BASE LOADDTM: DATE STUDYID: VARCHAR2(20) = 4488 FILCRDTM: DATE TRANCOMM: VARCHAR2(200)

«FK»+ FK_TRANSMISSION_LABORATORY(NUMBER)

«PK»+ PK_TRANSMISSION(NUMBER)

CDISC_UNKNOWNS

«column» SITEID: VARCHAR2(20) INVID: VARCHAR(20) INVNAM: VARCHAR2(80) SCRNNUM: VARCHAR2(20) SUBJSID: VARCHAR2(20) SUBJNIT: VARCHAR2(4) SEX: VARCHAR2(1) SEXCD: VARCHAR2(40) BRTHDTM: DATE RACE: VARCHAR2(20) RACECD: VARCHAR2(4 0) VISITMOD: VARCHAR2(20) ACCSNNUM: VARCHAR2(20) SPECNUM: VARCHAR2(10) PTMEL: VARCHAR2(9) PTMELTX: VARCHAR2(40) COLENDTM: DATE RCVDTM: VARCHAR2(25) SPECICOM: VARCHAR2(2048) AGEATCOL: NUMBER(3) AGEU: VARCHAR2(6) FASTSTAT: VARCHAR2(7) LBLOINC: VARCHAR2(10) LOINCCD: VARCHAR2(40) RPTRESCD: VARCHAR2(40) RPTRESNP: VARCHAR2(5) RPTNRLO: VARCHAR2(40) RPTNRHI: VARCHAR2(40) RPTUCD: VARCHAR2(40) CNVRESC: VARCHAR2(204 8) CNVRESCD: VARCHAR2(4 0) CNVRESN: VARCHAR2(20) CNVRESNP: VARCHAR2(5) CNVU: VARCHAR2(20) CNVUCD: VARCHAR2(4 0) SIRESC: VARCHAR2(204 8) SIRESCD: VARCHAR2(40) SIRESN: FLOAT SIRESNP: VARCHAR2(5) SINRLO: VARCHAR2(4 0) SINRHI: VARCHAR2(40) SIU: VARCHAR2(20) SIUCD: VARCHAR2(4 0) ALRTFL: VARCHAR(14) DELTFL: VARCHAR2(2) TOXGR: VARCHAR2(1) TOXGRCD: VARCHAR2(40) EXCLFL: VARCHAR2(14) BLNDFL: VARCHAR2(24) RPTDTM: VARCHAR2(25)

ELISA_RAW_ DATA

«column»*PK RAWDATID: NUMBER(10)*FK PLATEID: NUMBER(6) SPECMCD: VARCHAR2(40)*FK TESTID: NUMBER(6) WELLNUM: VARCHAR2(10) RPTRESC: VARCHAR2(2048) = NULL RAWVALU: FLOA T DILUTION: NUMBER(6) = NULL CALCVALU: FLOAT RPTRFLG: VARCHAR2(10)* TRANSMID: NUMBER(6)

«FK»+ FK_ELISA_RAW_DATA_ELISA_PLATE(NUMBER)+ FK_ELISA_RAW_DATA_TEST(NUMBER)

«PK»+ PK_ELISA_RAW_DATA(NUMBER)

ELISA_ STD_ DATA

«column»*PK STDDATID: NUMBER(6)*FK TESTID: NUMBER(6) STDNAM: VARCHAR2(40) FK PLATEID: NUMBER(6) WELLNUM: VARCHAR2(10) CONC: FLOAT CALCVALU: FLOAT RPTRESN: FLOAT = NULL CALCV: FLOAT CALSD: FLOAT RPTRESC: VARCHAR2(2048) = NULL RAWVALU: FLOA T RAWAVG: FLOA T RAWSD: FLOAT = NULL RAWCV: FLOAT = NULL* TRANSMID: NUMBER(6)

«FK»+ FK_ELISA_STD_DATA_ELISA_PLATE(NUMBER)+ FK_ELISA_STD_DATA_TEST(NUMBER)

«PK»+ PK_ELISA_STD_DATA(NUMBER)

TEST

«column»*PK TESTID: NUMBER(6) TSTCD: VARCHAR2(20) TSTNAM: VARCHAR2(100) RPTU: VARCHAR2(20) LBTEST: VARCHAR2(100) LBTESTCD: VARCHAR2(20)

«PK»+ PK_ELISA_TEST(NUMBER)

LABORATORY

«column»*PK LABID: NUMBER(5) TRMSRNUM: VARCHAR2(20) = ORTEL TRMSRNAM: VARCHAR2(40) = Duke Hemostasis... PLBNAM: VARCHAR2(40) = Duke Hemostasis... PLBNUM: VARCHAR2(20) = ORTEL LBNAM: VARCHAR2(40) = Duke Hemostasis... LBNUM: VARCHAR2(20) = ORTEL

«PK»+ PK_LABORATORY(NUMBER)

+FK_CLOTTING_DATA_TRANSMISSION

0..*(TRANSMID = TRANSMID)

«FK»

+PK_TRANSMISSION

1

+sample

1[SPECMCD = SPECMCD]+clottingDataCollection

0..*

+sample

1

+commentCollection

0..*

«table» CDISC_UNKNOWNS«flow»

«table» CDISC_UNKNOWNS

«flow»

+FK_ELISA_AVG_DATA_ELISA_PLATE

1..*(PLATEID = PLATEID)

«FK»+PK_ELISA_PLATE

1

+elisaPlate

1

+commentCollection

0..*

+testDat e 1

+commentCollection 0..1

+sample 1

+elisaDataCollection 0..*

+FK_ELISA_STD_DATA_ELISA_PLATE

1..*

(PLATEID = PLATEID)

«FK»

+PK_ELISA_PLATE

1

+FK_ELISA_RAW_DATA_TEST

0..*

(TESTID = TESTID)

«FK»

+PK_ELISA_TEST

1+FK_ELISA_AVG_DATA_TEST

0..*

(TESTID = TESTID)

«FK»

+PK_ELISA_TEST

1

+FK_ELISA_PLATE_TRANSMISSION 0..*

«FK»

+PK_TRANSMISSION 1

+FK_CLOTTING_DATA_TEST

0..*

(TESTID = TESTID)

«FK»

+PK_ELISA_TEST 1

+FK_TRANSMISSION_LABORATORY

0..*(LABID = LABID)

«FK»+PK_LABORATORY

1

+FK_SAMPLES_4488_TRANSMISSION

0..*

(TRANSMID = TRANSMID)

«FK»

+PK_TRANSMISSION

1

+FK_COMMENT_4488_TRANSMISSION

0..*

(TRANSMID = TRANSMID)

«FK»

+PK_TRANSMISSION

1

+FK_ELISA_STD_DATA_TEST

0..*

(TESTID = TESTID)

«FK»

+PK_ELISA_TEST

1

+FK_ELISA_RAW_DATA_ELISA_PLATE

0..*

(PLATEID = PLATEID)

«FK»

+PK_ELISA_PLATE

1

Design

BioSignatures BioSignatures \\ Biomarker StudiesBiomarker Studies

100% eSource

Metadata heavy

Sample management

New workflow

13

T1 Challenges in CDM Services ContextT1 Challenges in CDM Services ContextEvolving science causes evolving data Evolving science causes evolving data requirementsrequirements•• Scope, pricing, workflow, bio and information Scope, pricing, workflow, bio and information

science skillsscience skills

Discovery based work now within scope of Discovery based work now within scope of FDA FDA regsregs and and pharmapharma traditionstraditions•• Mismatch of burden/benefit and common Mismatch of burden/benefit and common

understanding of regulations understanding of regulations

Data Management systems are immature or Data Management systems are immature or don’t exist. don’t exist. •• Bioinformatics tools lack data management Bioinformatics tools lack data management

workflowworkflow•• IT IT --> Research IT> Research IT\\Computer ScienceComputer Science

Decision support methodology:Decision support methodology:implementation on a clinical trialimplementation on a clinical trial

Computer-assisted treatment recommendations

Web-based reports

Clinical Operations Clinical Operations reviews and reviews and

manages site manages site communicationscommunications

Site investigators

Decision Support

Tools

Study participants(patients)

Web-based Study Database

(21CFR Part 11 compliant)

Site contact information

Enrollment and randomization information

MD reviewer MD reviewer evaluates evaluates

and enters a and enters a Treatment Treatment DecisionDecision

Lab results

Treatment arm (as needed)

WilgusWilgus, et al. Poster presented at AMIA Annual Meeting, 2009. , et al. Poster presented at AMIA Annual Meeting, 2009.

14

T2 Challenges in CDM Services ContextT2 Challenges in CDM Services Context

New risk profile of research tasks within New risk profile of research tasks within patient care process. patient care process.

EHR’s EHR’s --> Disease cohorts > Disease cohorts --> Trials > Trials --> EHR> EHR•• Complex governanceComplex governance•• Consent; research Consent; research vsvs quality improvement; quality improvement;

future use; Identification.future use; Identification.•• Data collection design Data collection design

InteroperabilityInteroperability•• CDISC : HL7; WHO Drug : CDISC : HL7; WHO Drug : RxNormRxNorm; ;

MedDRAMedDRA : … ; Metadata!: … ; Metadata!

15

Duke CDR & Knowledge RepositoryDuke CDR & Knowledge Repository

Omics Data

CRF/Clinical

Data

CDR & Knowledge Repository

ImagingData

OperationalData

Study Meta Data

Electronic Health

Records

Cohort selection

Sample Data

Geospatial/Environmt.

Externalsources

Consent

Discovery Decisionsupport

Key IssuesKey IssuesUnanticipated acceleration of trendsUnanticipated acceleration of trends•• CTSA, ARRA, Duke Center for Health CTSA, ARRA, Duke Center for Health

Informatics, DCRI BioSignatures ProgramInformatics, DCRI BioSignatures Program

What is informatics? What is informatics? •• Depends on who you askDepends on who you ask•• Functional bounds of interdisciplinary domainFunctional bounds of interdisciplinary domain

Field is still immatureField is still immature•• Too few examplesToo few examples•• Talent gapTalent gap

Cost constraints & allocationCost constraints & allocation•• Project Project vsvs infrastructureinfrastructure

16

Where Next?Where Next?

Practical need for CRI will meet vision Practical need for CRI will meet vision (bottom up meets top down)(bottom up meets top down)

Grow as research partner, beyond service Grow as research partner, beyond service providerprovider•• CRI FacultyCRI Faculty

Data InfrastructureData Infrastructure•• Adopt HL7 Development Framework as Adopt HL7 Development Framework as

methodologymethodology•• Heavily use & contribute to standardsHeavily use & contribute to standards•• Leadership, data governanceLeadership, data governance

ConclusionsConclusions

The challenges discussed at this meeting The challenges discussed at this meeting have corresponding business challengeshave corresponding business challenges

The research trends driving the growth of The research trends driving the growth of CRI existCRI exist•• Not only as CTSA phenomena Not only as CTSA phenomena •• Can be expected to impact services marketCan be expected to impact services market

CRI is not yet well defined or establishedCRI is not yet well defined or established

Need to demonstrate valueNeed to demonstrate value

17

AcknowledgementsAcknowledgements

Connor Connor BlakeneyBlakeney

Robert Harrington, MDRobert Harrington, MD

Meredith Meredith NahmNahm

James James TchengTcheng, MD, MD

Swati Swati ChakrabortyChakraborty, , M.EngM.Eng..

Carol Hill, PhDCarol Hill, PhD

Cindy Cindy KlucharKluchar, MS, MS

James Topping, MSJames Topping, MS

Becky Becky WilgusWilgus, RN, MSN, RN, MSN

Evolution of a Clinical Research Evolution of a Clinical Research Informatics Group within a Informatics Group within a

ServiceService--oriented Clinical Trials oriented Clinical Trials Data Management OrganizationData Management Organization

B. McCourt, D. B. McCourt, D. FastesonFasteson--Harris, Harris, S. S. ChakrabortyChakraborty, C. , C. BovaBova HillHill

AMIA CRI SummitAMIA CRI SummitSan Francisco, March 2010San Francisco, March 2010