Building a CTRC Consortium Platform: Informatics, Data Sharing and Management of Clinical Trials in Specific Disease Areas Mike Conlon, University of Florida Paul Harris, Vanderbilt University
Feb 25, 2016
Building a CTRC Consortium Platform: Informatics, Data Sharing and Management
of Clinical Trials in Specific Disease Areas
Mike Conlon, University of FloridaPaul Harris, Vanderbilt University
The Challenge
• Clinical and Translational Research involves informatics beyond the needs of clinical care– Molecular Level information– Electronic Data Capture– Clinical Trials– Data warehousing– Data sharing– Information Discovery and Dissemination– Scientific Portfolio Management
Emerging Molecular Informatics
• Genetics– Personalized medicine – pharmacogenomics,
disease risk• Proteomics• Metabolomics
– Global and targeted– Mass Spectroscopy and Nuclear Magnetic
Resonance Imaging
Molecular Informatics at Vanderbilt
PRED
ICT:
Pha
rmac
ogen
omic
Res
ourc
e fo
r
Enha
nced
Dec
ision
s In
Care
and
Tre
atm
ent
Denny JC et al. PheWAS: demonstrating the feasibility of a phenome-wide scan to discover gene-disease associations. -- Bioinformatics 2010 May 1;26(9):1205-10.
Ritchie MD et al. Robust replication of genotype-phenotype associations across multiple diseases in an electronic medical record. -- Am J Hum Genet 2010 Apr 9;86(4):560-72.
Pulley J et al. Principles of human subjects protections applied in an opt-out, de-identified biobank. -- Clin Transl Sci 2010 Feb;3(1):42-8.
Smith JP et al. PGE2 decreases reactivity of human platelets by activating EP2 and EP4. -- Thromb Res 2010 Jul;126(1):e23-9.
Pulley J et al. Identifying unpredicted drug benefit through query of patient experiential knowledge: a proof of concept web-based system. -- Clin Transl Sci 2010 Jun;3(3):98-103.
Schildcrout JS et al. An analytical approach to characterize morbidity profile dissimilarity between distinct cohorts using electronic medical records. -- J Biomed Inform 2010 Dec;43(6):914-23.
Dumitrescu L et al. Assessing the accuracy of observer-reported ancestry in a biorepository linked to electronic medical records. -- Genet Med 2010 Oct;12(10):648-50.
Baye TM et al. Mapping genes that predict treatment outcome in admixed populations. -- Pharmacogenomics J 2010 Dec;10(6):465-77.
Denny JC et al. Identification of genomic predictors of atrioventricular conduction: using electronic medical records as a tool for genome science. -- Circulation 2010 Nov 16;122(20):2016-21.
Ramirez AH et al. Modulators of normal electrocardiographic intervals identified in a large electronic medical record. -- Heart Rhythm 2011 Feb;8(2):271-7.
Wilke RA. High-density lipoprotein (HDL) cholesterol: leveraging practice-based biobank cohorts to characterize clinical and genetic predictors of treatment outcome. -- Pharmacogenomics J 2011 Jun;11(3):162-73.
Malin B et al. Never too old for anonymity: a statistical standard for demographic data sharing via the HIPAA Privacy Rule. -- J Am Med Inform Assoc 2011 Jan 1;18(1):3-10.
Feng Q et al. A common CNR1 (cannabinoid receptor 1) haplotype attenuates the decrease in HDL cholesterol that typically accompanies weight gain. -- PLoS One 2010 Dec 31;5(12):e15779.
Wilke RA et al. The emerging role of electronic medical records in pharmacogenomics. -- Clin Pharmacol Ther 2011 Mar;89(3):379-86.
Turner SD et al. Knowledge-driven multi-locus analysis reveals gene-gene interactions influencing HDL cholesterol level in two independent EMR-linked biobanks. -- PLoS One 2011 May 11;6(5):e19586.
Higginbotham KS et al. A multistage association study identifies a breast cancer genetic locus at NCOA7. -- Cancer Res 2011 Jun 1;71(11):3881-8.
Xu H et al. Facilitating pharmacogenetic studies using electronic health records and natural-language processing: a case study of warfarin. -- J Am Med Inform Assoc 2011 Jul-Aug;18(4):387-91.
Wilke RA et al. Genetics and variable drug response. -- JAMA 2011 Jul 20;306(3):306-7.
Delaney JT et al. Predicting clopidogrel response using DNA samples linked to an electronic health record. -- Clin Pharmacol Ther 2012 Feb;91(2):257-63.
Personalized Medicine at Florida
Electronic Data Capture
• Strong need for simple methods for collecting data into electronic forms for support of clinical research
• Popular software is REDCap – used by a majority of CTSAs and many other institutions
REDCap Project History
2004 Needs AssessmentResearchers needed help managing data for small/medium sized non-trial research projects (pilot, R01, PPG)
HypothesisResearchers will do the right thing (secure, audit trails, etc) if provided an easy way to get needed tools
ProblemMany projects, few resources
REDCap Project History
Solution: Metadata-driven application (no per-project programming)
2004 - First REDCap project operational at Vanderbilt
2006 - REDCap ConsortiumLaunched REDCap Consortium to share with other universities and foster collaboration for future development
Case Report FormsVisual Status Data Validation
NumerousField Types
+ Text (Free) (Number) (Phone) (Zip) (Date)+TextArea+Select+Radio+File
BranchingLogicAuto-Variable Coding
HumanReadableLabels
PDFs
Data Export + De-ID Tools
Exports RawData + StatsScript Files(Labels, Coding
EmbeddedDe-IdentificationTools
Clinical Trial Management Systems
Functionality• All aspects of trial management• Time and event management• Electronic data capture• Recruitment support• Interface to clinical systems,
laboratory, imaging, prescribing, warehouse
• Financial management• Regulatory support• Interfaces to analytic software
Items to consider• Hundreds of systems
replaced with one• Hundreds of processes
replaced with dozens• Required flexibility for
innovation• Required agility for
innovation
Approach to CTMS• Vanderbilt
- No Single CTMS supporting research enterprise- StarBRITE for Recruitment, Regulatory, Financial and other CTMS
components. Heavy use of REDCap for electronic data capture.
• Florida– 1,000 new clinical studies per year– No CTMS, heavy use of REDCap, 100+ local systems in use for
trials
• Others– Velos– Oncore (especially in Cancer Domain)
Data Warehousing
• Data archive for cohort identification, trial planning, recruitment, registries
• Create data flows from clinical, laboratory, tissue bank and prescribing systems
• Create data flows from consent, trial, and molecular systems
• Researchers mine data for planning and results• Care managers mine data for planning and quality
improvement
Vanderbilt Data Warehousing
Participant Recruitment Example
Vanderbilt Data Warehousing
Snapshot – Pilot Studies
Nephrology• Examined: 2598• Candidates:
96 (reduction - 96%)Cleft Palate• Examined: 2490• Candidates:
27 (reduction - 99%)
Cardiology (2 studies)(reduction - 95%)
Starting Here Filtering Criteria
…
ReviewList
Clinics ofInterest
Study Work Queue (Daily Review)
Florida Data Warehousing
• Integrated Data Repository with joint governance by research and care
• Work began in 2011• Based on i2b2 software, with common data
model across studies, hospital and outpatient• All hospital data for 10 years, personalized
medicine
Integrated Data Repository
Data Sharing -- Layered
• Share data across institutions at various levels of aggregation – simple counts of procedures and diseases to full personal health records
• Technical considerations – definitions, data representation
• Policy considerations – risk management, privacy, competitive considerations
Layered Sharing
• Vanderbilt Institute for Clinical and Translational Research– Vanderbilt Medical Center– Meharry Medical Center
• Florida– Hospital on Jacksonville, 114 km. Epic software– Hospital in Orlando, 182 km. Crimson software– Hospital in Tampa, 209 km. Proposed SHRINE software
Data Sharing -- Study
• National Health and Nutrition Survey (NHANES)– Federal Effort– Longitidanal– Common data elements– Available for data mining
• Study Level– Data Use Agreements– Standardized definitions for measurement– De-identification
Information Discovery and Dissemination
• Need to know what is happening in science – papers, presentations, grants, datasets, funding, proposals, events
• Internet – portals, email, Facebook, Twitter• Local CTSA Example: Vanderbilt StarBRITE• VIVO – open software for research discovery
VIVO in China: http://health.las.ac.cn/
Scientific Portfolio Management
• Diversity, proportionality across the four translations
• Alignment with national, institutional and research strategic planning, goals and objectives
• Return on investment
Governance
• Joint (care enterprise, research enterprise) decision making– IT principles– IT Architecture– IT Infrastructure strategies– Application needs– IT investment and prioritization
Questions and Discussion