Objectives. The Goals of the TVT Registry

Transcription

1 STS/ACC TVT Registry From Beginning to Now Joan Michaels, RN, MSN, CPHQ, AACC March Objectives Discuss past and present STS/ACC TVT Registry activities Identify at least two benefits to participating in the TVT Registry 2 The Goals of the TVT Registry Learn from patient-level data Regulatory device surveillance Quality improvement Insights into patient selection, etc. Feedback, benchmarking, and best practices at a site level Patient education and informed decision-making Research important hypotheses tested to expand our understanding Be a driving force in improving our health care system 1

2 2002, Rouen, F.I.M TAVR Moving from concept to clinical execution Oh! my God! Day-8 July 2011 Nov 2011 Dec 2011 May 2012 Feb 2013 Sept 2013 Oct 25, 2013 Jan 17, 2014 May 18, 2014 May 2014 TVT Registry proposed at FDA Advisory Panel for Edwards Sapien Valve FDA Approves Edwards Sapien THV STS/ACC TVT Registry Launched Version 1.0 CMS issued National Coverage Decision STS/ACC TVT Registry Launched Version 1.2 STS/ACC TVT Registry Launched Version 1.3 FDA Approves Abbott MitraClip FDA Approves Medtronic CoreValve Anticipated Decision for NCD for MitraClip STS/ACC TVT Registry Launch for 2.0 Governance Structure of the TVT Registry 2

3 Members David Holmes Michael Mack John Rumsfeld Fred Edwards Eric Peterson Cynthia Shewan Kathleen Hewitt Members Fred Edwards (chair) Dave Cohen Eric Peterson Dave Shahian John Rumsfeld DCRI Project Team Sean O Brien Matthew Brennan Dadi Dai Michael Booth Risk Model Group The primary goal of the Work Group is to develop models (mortality, bleeding) used in TVT Registry Institutional Outcome Reports for benchmarking purposes. Other risk models, such as one estimating individual patient risk may follow. The initial model compares in-hospital mortality across participating centers, potentially identifying outliers as more data accumulate in the Registry. All 13,218 Cases are Commercial TAVR 272 Sites Enrolled in TVT Registry 258 Sites Entering Data 5 Sites > 200 Cases; 24 Sites > 100 Cases Does Not Include All TAVR Cases in US Over 10K Cases from IDE Trials Some Commercial TAVR Cases Before NCD, VA Cases 3

4 Distribution of Sites by Annual Volume Number of Sites Note: Investigative case volume at sites isnot counted Annualized Volume Groups Source: STS/ACC TVT Registry Database Sites Participating in TVT Registry (Currently TAVR) No TAVR sites in Alaska and Wyoming (currently patients served by Seattle and Colorado sites) 1. As of last week 272 TAVR sites fully enrolled. Additional sites being trained by Edwards and Medtronic will be added. 2. Approximately 35 sites who have started commercial MitraClip use are already TAVR-TVT sites. STS/ACC TVT Registry Data Quality Program 4

5 Data Quality Program Pre data collection Data submission Post data submission Software Training Clinical support Data quality reports Adjudication Audit Outlier program Publications Data Quality Program Post Submission Adjudication Verifies and provides additional information for key events (stroke, TIA and repeat intervention, plus CHF admission for MitraClip) National audit program (starts in 2014) Evaluates accuracy and reliability Assesses proper and complete reporting of cases Voluntary and self audits Data Outlier Program Provides outlier alerts to Registry participants Adjudication Process: Additional Data and Physician Review at DCRI 5

6 KCCQ Data Completeness Update Improvements & Expansion in Data Elements and Modular Design Version Reduction of site burden in TAVR data collection based on lessons learned 2. Expansion to include CoreValve technology 3. Development of Mitral Module to include MitraClip technology 4. Formation of Three Separate TVT Registry Modules STS Members Fred Edwards Patrick McCarthy Craig Miller Echo Specialists Paul Grayburn *** Neil Weissman TVT Registry Mitral Module Workgroup Composition ACC Members John Carroll - Chair Samir Kapadia Data Analytic Center (DCRI) FDA - John Laschinger*** STS, DCRI and ACC Staff Cynthia Shewan and Donna McDonald STS Michael Booth and Paul Meehan - DCRI Joan Michaels and Susan Fitzgerald ACC *** Heart Failure Specialists Maggie Redfield JoAnn Lindenfeld FDA CMS Stakeholders Industry ACC and STS leadership Analytic Centers Users (public comment period) Abbott Input Barathi Sethuraman Bhagya Gurusu Liz McDermott 6

7 Mitral Module and CMS CMS held an open comment period for a National Coverage Decision for MitraClip procedures NCD expected May 2014 Data Elements and Definitions Aligned with VARC2 endpoints Harmonized with STS and ACC-NCDR data elements where possible Data elements and definitions reviewed and approved by Industry, CMS and FDA 8 page DCF Health Status (KCCQ) Six Minute Walk 30 day and 1 year follow-up Adjudication of TIA/Stroke Re-interventions Heart failure readmission Data Elements and Definitions 7

8 Reporting from the TVT Registry Presentations at Major Meetings: TCT, STS, CRT, ACC, etc. Abstracts Manuscripts Reporting on TVT Registry Data But Also Demonstrating the Added Impact from Links to Other Registries and Databases ACC 2014 Late-Breaking Clinical Trial: One Year Outcomes from the STS/ACC Transcatheter Valve Therapy (TVT) Registry. David R. Holmes, J. Matthew Brennan, John Rumsfeld, Dadi (David) Dai, Fred Edwards, John Carroll, David Shahian, Frederick Grover, E. Murat Tuzcu, Eric Peterson, Ralph Brindis, Michael Mack. CMS has successfully linked the STS/ACC TVT Registry to inpatient Medicare claims using direct patient identifiers with a roughly 97% record linkage rate across more than 7,500 TAVR cases from Nov 2011 through October Additional Abstract Presentations 8

9 Completeness of Data DCRI Analysis 9/13/2013 n = 7797 cases Data Fields: Range of Complete data Pre-Procedure Status Clinical Data(labs, PFTs): 10 elements 43%- 100% DLCO 43% Pre-Procedure Meds: 5 elements % Pre-Procedure Status: Cath and Echo Data: 30 elements % Left atrial area 32% Improvements in TAVR Module in Version 2.0 Reduction in number of data elements Current version = 343 Version 2.0 = 308 with 38 data elements eliminated Added UDI field Added fields to assess disability from stroke without requiring burdensome Rankin score Social/Recreational Activities Impaired Neurocognitive Functions Essential to Patient Impaired New Aids or Assistance Required Who is Undergoing TAVR in the USA? Demographics: Age, Sex, Race, Ethnicity The Baby Boomers are Coming and US Demographics are Shifting in Race/Ethnicity 9

10 Women What Does the STS-ACC TVT Registry Show? (N= 7710 Patient Records) Expected Percentage of TAVR Patients Who are Predicted to be Women Based on US Census, TAVR Age Data, and Assumed Disease Frequency 61.9% WHY the difference? Actual Percentage of TAVR Patients Who are Women 49.1% Black-African/Americans What Does the STS-ACC TVT Registry Show? 7710 Patient Records Expected Percentage of TAVR Patients Who are Predicted to be Black Based on US Census, TAVR Age Data, and Assumed Disease Frequency 9% WHY the difference? Actual Percentage of TAVR Patients Who are Black 3.6% Research and Publications Committee STS Members Fred Edwards Vinod Thourani Jeffrey Jacobs Todd Dewey ACC Members John Rumsfeld David Cohen Neal Kleiman Ajay Kirtane Registry Operations Center NCDR Ex Officio Members FDA Art Sedrakyan Biostats - Eugene Blackstone DCRI Eric Peterson Staff Tina Kourtis Mary Weideman, Joan Michaels, Kathleen Hewitt, 10

11 Research and Publications Initial Phase: Steering Committee Members to Provide the First Wave of Strategic Presentations and Publications Current Phase: January 2014 the R & P Committee accepted outside research proposals for using the TVT Registry data 80 proposals were received! R & P committee now performing reviews and establishing priorities Post-Approval Studies, IDE s, and Uses of Registry Data for Expansion of Indications STS-ACC actively involved with discussions with FDA How are data collected in observational registries? How can data be used to support post-approval studies and investigative device exemptions ( IDE s )? What are the applicable informed consent requirements? How to best utilize the vast amount of patient-level data while assuring the safety and protection of human subjects and imposing the least burden on the sponsors and participating hospitals? Outcomes from TAVR and Surgical AVR Is Consumer Level Reporting Coming? 11

12 Outcomes of TAVR at Different Sites How and when can the TVT registry provide an analysis of outcomes per site that is credible and actionable? 1. To assist patient, families, and clinicians in site selection 2. To provide CMS, professional societies, and medical device companies on the success of the current site qualification standards used in the NCD 3. To provide guidance on whether the number of TAVR sites should be further expanded with less stringent qualifications 4. To determine learning curve effects, the volume-outcome relationship in TAVR, and the determination of best practices to improve outcomes 5. To start a dialogue on how sites with poor outcomes should be approached. Step #1: Outcomes Must Be Adjusted to The Characteristics of the Patients Being Treated Is the observed outcome equal, less, or more than predicted for a site s patients? The risk-model development group: Fred Edwards and others The first TVTR risk model should be available in May Other Factors Determining Outcomes TAVR technology is changing and the TVT Registry should show substantial changes (reductions) of many complications as new and better valves/delivery systems are approved. The learning curve continues to operate for the whole community and individual sites. Identification of patients who may have underlying characteristics predicting a poor outcome may help guide centers in the key issue of patient selection 12

13 Other Lessons Learned in Judging Sites Statistical assessment of sites with low volumes is problematic. Geographical distribution of sites is key for patient access Assuming 1. Patients, Families, and Clinicians Have Access to TAVR and savr 2. Sites Have Outcomes That Are at Least as Good as Expected How Can The TVT Registry Be Used to Help Patients, Families, and Clinicians in Key Patient-Centric Decisions? Optimizing Health Outcomes in Patients with Symptomatic Aortic Valve Disease. PICORI Grant Awarded Late 2013 Matthew Brennan, MD Principle Investigator DCRI 13

14 Research Goal: Help patients make informed treatment decisionsthat incorporate their preferences and outcomes of interest 1. Creation and assessment of statistical prediction models accurate and valid patient-level predictions of outcomes for patients with severe aortic stenosis treated with SAVR vs. TAVR. 2. Creation of decision assistance tools suitable for use by patients Surgical candidates A second tool for patients who have surgically inoperable disease. 3. Dissemination of these tools into clinical practice using a webbased platform coupled with a pilot evaluation of their effectiveness Questions? 14