1. Collaborative Networks for Conducting Comparative Effectiveness Research Tuesday September 9, 2008 8:00 – 9:30 am

2. Session Outline 8:00 – 8:15 AHRQ introduction – George Neyarapally 8:00 – 8:15 AHRQ introduction – George Neyarapally 8:15 – 8:50 CO Distributed Research Network (DRN) project – Wilson Pace, David West 8:15 – 8:50 CO Distributed Research Network (DRN) project – Wilson Pace, David West 8:50 – 9:25 HMORN Distributed Research Network (DRN) project – Rich Platt, David Magid 8:50 – 9:25 HMORN Distributed Research Network (DRN) project – Rich Platt, David Magid 9:25 – 9:30 Questions 9:25 – 9:30 Questions

3. Introduction: AHRQ’s Distributed Research Network Initiative in DEcIDE George Neyarapally PharmD, MPH, RPh AHRQ, Center for Outcomes and Evidence 3

4. Outline Background Background Rationale for Distributed Research Network Initiative Rationale for Distributed Research Network Initiative Goals of the Task Order Goals of the Task Order Benchmarks of Success Benchmarks of Success

5. DEcIDE Research Network Established as part of MMA in 2005 and incorporated in the Effective Health Care program. Established as part of MMA in 2005 and incorporated in the Effective Health Care program. 13-center research network that is designed to be a data and scientific resource to external public & private health organizations (particularly CMS) to provide unbiased, actionable information to inform decision makers. 13-center research network that is designed to be a data and scientific resource to external public & private health organizations (particularly CMS) to provide unbiased, actionable information to inform decision makers. Main goal of DEcIDE is to develop new knowledge about the clinical and comparative effectiveness and safety of treatments and technologies in a valid, transparent, and scientifically rigorous manner. Main goal of DEcIDE is to develop new knowledge about the clinical and comparative effectiveness and safety of treatments and technologies in a valid, transparent, and scientifically rigorous manner.

6. Distributed Research Network Prototypes for Population-based Studies Two pilot project models of distributed research networks will be created & evaluated. Two pilot project models of distributed research networks will be created & evaluated. The aim is to develop a distributed network prototype that supports secure analyses of electronic information across multiple organizations to study risks, effects, and outcomes of various medical therapies. The aim is to develop a distributed network prototype that supports secure analyses of electronic information across multiple organizations to study risks, effects, and outcomes of various medical therapies. The long-term goal is a coordinated partnership of multiple research networks that provide information that can be quickly queried and analyzed. The long-term goal is a coordinated partnership of multiple research networks that provide information that can be quickly queried and analyzed.

7. Unique Aspects of Effective Health Care Program User-driven research & systematic reviews. User-driven research & systematic reviews. Transparent methods, with process for reviewing COI with every project. Transparent methods, with process for reviewing COI with every project. Research for public health: Research for public health: – Focused on improving quality of Medicaid, Medicare, SCHIP. – Programs with rising costs. Partnership with stakeholders & public nomination. Partnership with stakeholders & public nomination.

8. Rationale for Task Order Silo investigators and databases. Lack of coordination of research across multiple institutions. Silo investigators and databases. Lack of coordination of research across multiple institutions. Researchers and decision-makers are disconnected, often working independent of each other. Researchers and decision-makers are disconnected, often working independent of each other. Small studies often with limited external validity / generalizability which constrains good decision making. Small studies often with limited external validity / generalizability which constrains good decision making. Vast collections of data that are infrequently shared. Vast collections of data that are infrequently shared. Length to study completion does not meet the needs of policymakers, providers or patients. Length to study completion does not meet the needs of policymakers, providers or patients. Perceived inability of government to respond quickly to emerging clinical questions & public health problems that face regulators, payers, providers, & individual patients. Perceived inability of government to respond quickly to emerging clinical questions & public health problems that face regulators, payers, providers, & individual patients. Current Research Paradigm?

9. Rationale for AHRQ’s Distributed Research Network Initiative Leverage existing resources including intellectual (scientists) and data resources. Leverage existing resources including intellectual (scientists) and data resources. Create a pipeline for conducting rapid cycle research related to clinical effectiveness and comparative effectiveness. Create a pipeline for conducting rapid cycle research related to clinical effectiveness and comparative effectiveness. Reduce the timeline for providing valid evidence that can inform clinical decisions and public health problems. Reduce the timeline for providing valid evidence that can inform clinical decisions and public health problems. Prepare for the future needs of federal government to address questions of public policy in real time. Prepare for the future needs of federal government to address questions of public policy in real time.

10. Examples Simple: Simple: – How many patients receive carotid stents as compared to medical management? More complex: More complex: – Are patients who take dual therapy with ACE inhibitors and angiotensin receptor blockers at increased risk for adverse outcomes? – What is the balance of risks and benefits of combination therapy?

11. Major Components of AHRQ’s DRNI Two pilot projects Two pilot projects – HMORN – University of Colorado & AAFP Internal working group including HIT staff and coordination with FDA and other federal health agencies. Internal working group including HIT staff and coordination with FDA and other federal health agencies.

12. General Goals of RFTO Develop a scalable, evolvable, model that initially differentiates itself from other distributed networks by its capability to support user-driven research on the safety, clinical effectiveness, and comparative effectiveness of treatments. Develop a scalable, evolvable, model that initially differentiates itself from other distributed networks by its capability to support user-driven research on the safety, clinical effectiveness, and comparative effectiveness of treatments. Develop a research cooperative and governance structure that advances the traditional model to assist decision makers. Develop a research cooperative and governance structure that advances the traditional model to assist decision makers. – This advancement is essential for the continuation of public trust and funding from public and private payers.

13. General Goals of RFTO Understand the extent to which the distributed research paradigm can be applied to problems facing policymakers, clinicians, & patients. Understand the extent to which the distributed research paradigm can be applied to problems facing policymakers, clinicians, & patients. – Where does the model exist in the continuum of research? Establish program governance. Establish program governance. – Not only address the technical aspects but the “sociologic” aspects of a research cooperative – Minimize barriers to collaboration and close coordination with scientific staff at AHRQ. – Relationship building.

14. Four Phases of Projects Technical Specifications – Spring to Fall 2008 Technical Specifications – Spring to Fall 2008 Pilot testing – Spring 2008 to Winter 2008 Pilot testing – Spring 2008 to Winter 2008 Research Study – Summer 2008 to early 2009 Research Study – Summer 2008 to early 2009 Blueprint Prospectus – Spring 2009 Blueprint Prospectus – Spring 2009

15. Elements of the Ideal System Create an open architecture that is technically and socially scalable to extend participation and collaboration with other AHRQ programs to meet the MMA mandates. Create an open architecture that is technically and socially scalable to extend participation and collaboration with other AHRQ programs to meet the MMA mandates. Develop a system which allows timely access to real time information and ensure a rapid response to questions facing clinicians, patients, and policy makers. Develop a system which allows timely access to real time information and ensure a rapid response to questions facing clinicians, patients, and policy makers. Make technical and procedural burdens as light as possible without sacrificing scientific integrity. Make technical and procedural burdens as light as possible without sacrificing scientific integrity. Closer connection between researchers and decision makers. Closer connection between researchers and decision makers. Low-cost, lightweight system that requires modest infrastructure, easy maintenance, and a coordinated research agenda. Low-cost, lightweight system that requires modest infrastructure, easy maintenance, and a coordinated research agenda.

16. Short-term Benchmarks Provide new information that helps advance a distributed model in health care research. Provide new information that helps advance a distributed model in health care research. Enrich AHRQ’s current capacity to conduct observational research on CE, specifically in DEcIDE & CERTs. Enrich AHRQ’s current capacity to conduct observational research on CE, specifically in DEcIDE & CERTs. Form a close partnership with AHRQ and stakeholders so that useful information is derived from this large investment of resources. Form a close partnership with AHRQ and stakeholders so that useful information is derived from this large investment of resources.

17. DRN Projects 2.0 Future DRN projects – how will we build upon these pilot projects? Future DRN projects – how will we build upon these pilot projects?