Happy Sweet 16 months PCORnet! Update on Progress & Promise - PowerPoint PPT Presentation

Happy Sweet 16 months PCORnet! Update on Progress & Promise Sarah Greene, MPH, Associate Director CER Methods and Infrastructure Program, PCORI Health Datapalooza 2015

PCOR...what? Pink Cornet? No, PCORnet! P atient- C entered O utcomes R esearch net work PCORnet is PCORI’s flagship initiative to build an efficient and patient-centered platform for observational and interventional research Why? Hasn’t this ground already been covered?

Challenge-tunity: a reusable national infrastructure for clinical research Clinical trial coordinating centers, large research networks are often established for one purpose or one funding cycle  New trial or new condition often begets construction of new infrastructure Persistent inefficiencies in the current research process, from regulatory and operational, to recruitment and data collection hinder the research enterprise Most research doesn’t harness the full potential of electronic health record (EHR) data

PCORnet: Big Data, Easier Processes, Co-Creation by Stakeholders A data platform is needed to support PCORnet, but we also need the data to have utility in real-world healthcare Through PCORnet, we will have ongoing input from patients, clinicians, health system leaders, payers and other stakeholders on priority research topics Engagement is the “special sauce” that will enable PCORnet to provide the answers patients need more quickly and efficiently, and at lower unit cost, than has ever been possible

Vision for PCORnet: enable rapid, large-scale, patient-centered clinical research in our healthcare systems and communities PCORnet is about PCORI is about Research Research Infrastructure Done Differently Done Differently Engagement is the cornerstone

PCORnet Facts & Figures 29 networks + Coordinating Center  11 Clinical Data Research Networks Complementary and synergistic capabilities in the 2 types of networks  18 Patient-Powered Research Networks  Of the 18, nine are focused on rare diseases 155 involved organizations across the US 3000 + collaborators/contributors Millions of patients receive care in the participating systems Phase I = March 1, 2014 – September 30, 2015 Phase II = October 1, 2015 – September 30, 2018 ≈ $275M: PCORI investment in infrastructure and initial projects 7

Diverse Sources + Rich Data = High Potential Federally Academic Disease Qualified Health Advocacy Health Centers Groups Centers Clinical & Integrated Patient- Health Translational Delivery Powered Information Science Systems Registries Exchanges Awardees mHealth / Electronic Pharmacy Data from Patient- Biospecimen Health Data Payers Generated Data Records Vendors (e.g., CMS) Data

Rapidly Growing Data Capabilities PCORnet Common Data Model builds on prior data models (HMORN Virtual Data Warehouse, FDA Sentinel) Complementary to and compatible with both OMOP and i2b2 Most networks have harmonized local data to CDM Version 1.0 16 months in, we just ratified Version 3.0 of PCORnet CDM Readying data to be “interrogated” for various research questions PopMedNet will be used as the secure querying engine  10 of 11 CDRNs and 8 of 18 PPRNs established on PopMedNet 9

Rather than bringing the data to the question, PCORnet is designed to bring the question to the data PCORnet DRN Coordinating Center 1 6 1. User creates and PCORnet Secure Network Portal submits query (a computer program) CDRN 1 2. Individual Review & Review & CDRNs/PPRNs Run Query Return retrieve query Results 2 5 3 4 Demographics 3. CDRNs/PPRNs Utilization Etc review and run query against their local data CDRN 11 4. CDRNs/PPRNs review results Review & Review & Run Query Return Results 5. CDRNs/PPRNs 3 4 return results via Demographics Utilization secure network Etc 6. Results are aggregated

Guiding principle: make research easier Analysis ready data  Standard format  Harmonized definitions  Quality checked in advance Reusable analysis tools Efficient clinical trial enrollment and follow up mechanisms Simple, pragmatic studies integrated into routine care Administrative simplicity (e.g., streamlined contracting) Initial PCORnet demonstration projects will help assess end-to-end functionality: design, implementation, analysis, and reporting

Demonstration projects will move PCORnet from concept to action Aspirin Clinical Trial (ADAPTABLE) Obesity Observational Studies  Bariatric Surgery Outcomes  Antibiotics and Weight Gain in Children PPRN Research Demonstration Projects Health Systems Demonstration Project NEXT-D Initiative with CDC Discussing topics of interest to health plans

Goals of PCORnet demonstration studies ADDRESS questions important to patients and clinicians that require multi-site evaluation ASSESS PCORnet’s ability to perform large-scale interventional and observational research FACILITATE collaboration between networks GUIDE further development of policies, procedures, infrastructure ROAD-TEST data and networking capabilities ENSURE PCORnet’s privacy protecting data infrastructure and analysis capabilities are sound DEVELOP efficient methods for identifying, enrolling, and following potential clinical trial participants

“Triple Aim” for Research: broad applicability and appeal to researchers, funders, health systems, payers, and patients Interventional Trials PCORnet Goal: Capacities in place to support all three types of research Rapid Cycle Observational Research on Studies Healthcare Delivery

Aspirin Dosing: A Patient-Centric Trial Assessing Benefits and Long-term Effectiveness (ADAPTABLE) Trial PCORnet’s First Pragmatic Clinical Trial

Illustrating the Need for an Aspirin Trial: Case Scenario Ted had chest pain while working and was taken to the ER where he learned he was having a heart attack. Doctors told him that plaque was building up in his arteries. Upon discharge from the hospital, Ted was advised to take 325mg of aspirin each day (regular strength). Ted compared notes with a friend who said his doctor has him on a baby aspirin because it causes less bleeding and bruising. Ted is confused about what dose he should take. He does a lot of work outdoors and carpentry. He is worried about bleeding while working but doesn’t want another heart attack either. And now Ted wonders what he should do. 17

Distribution of aspirin dosing at time of hospital discharge has been shown to vary Other 0.01% 81 mg 81 mg 36% 162 mg 325 mg 325 mg 61% other 162 mg 3% Hall et al. Circulation Cardiovascular Quality and Outcomes 2014

Aspirin: “wonder” drug in many ways Proven clinical benefit in reducing ischemic vascular events Cost-effective Benefit with combination antiplatelet therapies But there are issues:  Emerging evidence for dose modifiers (e.g., ASA resistance, genetics)  Equal efficacy across patients?  Intolerance Most effective dose uncertain— ADAPTABLE Trial aims to resolve that uncertainty

Trial Logistics: Taking Advantage of PCORnet Infrastructure 20

Study Goal n = 20.000 design Patients with known coronary artery disease (MI, or CAD or Revasc) + ≥1 “ enrichment factor”* Identified through EHR /direct patient consenting in CDRN & PPRN clinics/hospitals (PPRN patients will need to connect through a CDRN to participate) Pts. contacted electronically with trial information and eConsent ; treatment assignment will be provided directly to patient ASA 81 mg QD ASA 325 mg QD *Enrichment factors • age > 65 years • creatinine > 1.5 Electronic follow-up quex at 4 months; • diabetes supplemented with EHR/CDM/claims data • known 3-vessel CAD • current CVD &/or PAD • known ejection Duration: Enrollment over 24 months; fraction <50% maximum follow-up of 30 months • current smoker Primary Endpoint: Composite of all-cause mortality, nonfatal MI, nonfatal stroke Primary Safety Endpoint: Major bleeding complications

Use EHR data to create “computable phenotype” that can identify patients at participating CDRNs At least one of the following: History of CAD • age > 65 years • Past MI • Creatinine > 1.5 OR • Diabetes, • Past cath showing • Known 3 vessel coronary artery significant CAD disease OR • Current cerebrovascular • Revascularization disease and/or peripheral artery (PCI/CABG) disease • Known ejection fraction <50%, • Current smoker Obtain consent, electronically when possible 22

Centralized follow-up will also be electronic whenever possible OR HeH FOLLOW-UP DCRI FOLLOW-UP • Patient Reported Outcomes • Patient Reported Outcomes • Medication use • Medication use • Health outcomes • Health outcomes ADAPTABLE 12 16 20 …. 30 8 4 Enrollee PCORNet Coordinating Center FOLLOW-UP Baseline Data • Via Common Data Model • Longitudinal health outcomes CMS Virtual Data Warehouse FOLLOW-UP • Longitudinal health outcomes 23