Ralph G' Brindis, MD, FACC

1
NCDR Physicians Leading the Effort To Quantify
Quality CVS.42 Quality Improvement Initiatives
in Cardiology
NationalCardiovascularDataRegistry

• Ralph G. Brindis, MD, FACC
• Chief Medical Officer
• ACC-NCDR
• November 4, 2007

2
Disclosure Information
NCDR Physicians Leading the Effort To Quantify
Quality Ralph Brindis, MD, MPH, FACC, FSACI
Grant support (GS), consultant (C), speakers
bureau (SB), stock options (SO), equity interest
(EI)NONEOff label use of products will (not)
be discussed in this presentation NONE
3
Mission of the NCDR
To improve the quality of cardiovascular patient
care by providing information, knowledge and
tools implementing quality initiatives and
supporting research that improves patient care
and outcomes.
4
NCDR is
Ped. Registry
Achieve
Building a true
National CardioVascularDataRegistry
ICD Long
EP Registry
Imaging Registry
PracMgt Registry
Congenital Registry
PAD Registry
IC3 CAD
ACTION Registry
HF Registry
CARE Registry
CathPCI Registry
ICD Registry
1998.. 2004 2005 2006 2007 2008 beyond
5
NCDR Management Board
Advisory Council Industry Federal Health
Plans Patients
NCDR Operations Leadership Team
Scientific Oversight Committee
Data Safety Monitoring Board
Registry Projects
Registries
QI Projects
ambulatory
Committee structure for each registry
ACTION Registry
IC3 Program Steering Committee
Take ACTION Campaign Planning Work Group
Steering Committee
longitudinal
CathPCI Registry
ACHIEVE Registry Steering Committee
NCDR-D2B Project Managed by ACTION and CathPCI
Steering Committees
Research Publications
longitudinal
ICD Registry
Clinical Support.Team
ICD Longitudinal Program Steering Committee
QualityKIT/ CathKIT TBD
QI Subcommittee
CARE Registry
Includes 30day outcomes
Version 10/29/07
6
Executive Summary Page
7
CathPCI Report Executive Summary
Your Hospital
Your Hospital
Indicator
Rank percentile
Detail Line Number
Rank
Median
25 Pctl
Best Practice
75 Pctl
90 Pctl
10 Pctl
8

• Registry/QI
• gt985 hospitals
• 6 million patient records
• 2 millions PCI records
• Online data entry tool
• Support D2B Alliance
• Analytic Reporting Services
• States MA, OH, WV, ?CT, ?NJ
• Payers United, BCBSA, WellPoint
• Research and Publications
• DCRI analytic center
• Over 100 publications

9

• Registry
• 1425 enrolled
• 200,000 patient records
• Analytic Reporting Services
• UHC
• Discussions with BCBSA
• Provide data to CMS for reimbursement
• Research
• Abstracts at AHA
• ICD Longitudinal Study
• Performing analysis for FDA

10

• Registry
• 235 Participants
• gt 3,000 patient records
• Data entry tool
• CMS data requirement
• Research
• Analysis for FDA
• Discussion with industry - PMS

11

• Registry
• 300 participants
• Over 30,000 records by 9/07
• Funding provided by
• Genentech
• Bristol-Myers Squibb/Sanofi Partnership
• Schering Plough Corporation
• Analytic Reporting Services
• Early discussions with payers

12
The Cycle of Clinical Therapeutic Effectiveness
Clinical Trials
Concept
Guidelines
Outcomes
QUALITY
Performance Indicators
NCDR ICD, ACTION, CARE, CathPCI STS
Performance
13
Benchmarking Primary PCI lt90 Minutes
14
2004 STEMI ACC/AHA Guideline Update JCAHO Core
Measure
D2B Alliance Launch
15
ACC-Quality/CathKIT
CQI Tutorial Meeting Standards Reporting
Outcomes Implementing CQI
16
(No Transcript)
17
Hospital PCI Volume and In-Hospital
MortalityACC-NCDR 2001-2004

• Hospital PCI STEMI
  Non-STEMI Elective
• Volume (pts) n90,256 pts
  n94,587 pts n482,960 pts
• 200 vs gt800 0.99 (0.75,1.31)
  0.64 (0.38,1.06) 1.17 (0.81,1.71)
  
• 201-400 vs gt800 0.96 (0.83,1.12)
  0.87 (0.68, 1.10) 1.12 (0.96, 1.31)
• 401-800 vs gt800 0.95 (0.85,1.07)
  0.96 (0.81,1.14) 1.10 (0.99,1.22)
• Mortality 4.83
  2.09 0.41

(Odds Ratio, 95 CI)
Zhang et al Circulation 2005 Suppl II112792.
18
Performing Percutaneous Coronary Interventions at
Facilities Without On-Site Cardiac Surgical
Backup is IncreasingA Report FromThe American
College of Cardiology - National Cardiovascular
Data Registry

• Dehmer GJ, et.al. Am J Cardiol 200799329-332.

19
Proportion of Urgent PCIs with and
withoutOn-site Surgical Back-up
Jan 2001
Dec 2004
Calendar Quarter
20
Proportion of Elective PCIs with and without
On-site Surgical Backup
Jan 2001
Dec 2004
Calendar Quarter
21
PCI With or Without Onsite Surgery
StandbyACC-NCDR 2001-2004

• In-hospital Mortality Offsite vs Onsite CVSx
• Mortality Odds Ratio
  95 CI P-value
• No Acute MI (n482,018) 0.54 vs 0.41
  1.04 (0.67,1.62) 0.87
• STEMI (n 90,050) 4.65 vs 4.83
  0.96 (0.72,1.26) 0.75
• NSTEMI (n94,347) 1.94 vs
  2.09 0.67 (0.40,1.11)
  0.12

22
PCI With or Without Onsite Surgery
StandbyACC-NCDR January 2004 - March 2006

• 404 centers with Surgical Back-up
• 61 centers without Surgical Back-up
• 299,132 pts from centers with SOS
• 9,029 pts from centers without SOS
• 13 of Registry PCI patients
• Data verified via Quality Initiative Query

23
PCI With or Without Onsite Surgery
StandbyACC-NCDR January 2004-March 2006

• Unadjusted and Risk Adjusted Mortality
• Emergency CABG rate and CABG Mortality
• Elective and Emergent PCI
• Procedural success
• Door to Balloon times
• Descriptors of care
• PCI Volume, distance/time/mode of travel for off
  site Surgery, hospital characteristics, lesion
  risk, clinical variables for risk adjustment

24
Improving Continuous Cardiac CareOffice-Based
Registry
25
Improving Continuous Cardiac Care In the Office

• The first CAD office-based registry
• assess physician adherence to ACC/AHA clinical
  practice guidelines
• includes patients with Hx of ACS, prior PCI
  and/or CABG.
• Powerful tool that allows MD/Payer to assess and
  improve current office-based clinical care.

26
Philosophy of the IC3 Program

• Make it easier for busy clinicians to do the
  right thing for the right patient at the right
  time
• Track key performance measures
• Internal QI and P4P reporting at the practice
  level
• Make care more efficient
• A worksheet that readily identifies opportunities
  to apply CAD guideline recommendations and
  performance measures
• Coordinate care
• Create a visit summary to communicate with
  patients and other providers

27
Measuring CAD Care
AMI Care
Post-Hospitalization Risk factor
modification Cardiac rehabilitation
Patient with stable angina
Onset of Acute Coronary Syndrome
D/C
PCI/CABG
Admit
28
The IC3 Registry
Data Entered through NCDR IC3
Data entered and Clinic Visit Form Generated
Treatment plan Data entered
Pt presents for visit, reports med changes
Vitals, health status assessed
Physician Visit Rx
Patient Letter Visit Summary dispensed
Visit Summary sent to other care providers
29
IC3 Program Goals

• Provide QI tools designed for the entire
  office-based clinical care team
• Create QI tools directed at patients to become
  active participants and advocates for their own
  healthcare
• Explore strategies to support continuity of care
  among the multiple providers caring for an
  individual patient
• Provide real-time reporting of office-based
  quality indicators derived from clinical practice
  guidelines recommendations

30
IC3 Program Goals

• Create a trusted mechanism for measuring
  performance
• Serve as a valuable resource for research aimed
  at improving the treatment and outcomes of
  ACS/CAD patients in an ambulatory setting
• Support evolving CMS outpatient quality measures
  and regulatory reporting initiatives
• Support Pay-for-Performance programs

31
Sample QI Strategies

• Patient education resources
• Overview of ACS/CAD
• Explanation of treatment recommendations
• Visit-based summaries of treatment plans
• Printable versions for patients
• Encourage physician to physician communication
• Office identification and tracking systems
• Dissemination of best practices Health status
  tools and reporting features

32
ACCs Appropriateness Criteria
SPECT-MPICardiac CTCardiac MRIEcho TTE/TEE
Stress Coronary Revascularization PCI/CABG
33
(No Transcript)
34
64 Slice Coronary CT
35
Tools for Achieving Quality in Imaging
Lab accreditation Physician training Physician
competency
Appropriateness criteria Benchmarking Provider
education
Registries Research
Key data elements Uniform structured
reports Timeliness standards
Lab accreditation Technologist cert.
ACC-Duke Think Tank 2006 JACC 2006 48 2141
36
Pilot StudyEvaluation of Appropriatenessof
SPECT MPI

• The American College of Cardiology
• The American Society of Nuclear Cardiology
• NCDR

37
Purpose of the Project

• Facilitate quality improvements
• Efficient, effective patient care
• Evaluate promote awareness of appropriateness
  criteria in practice
• Provide feedback reports to improve both
  practice-level and individual physician-level
  adherence to the criteria
• Establish benchmarks to guide performance
  improvement
• Provide an alternative to prior authorization

38
SPECT MPI Appropriateness CriteriaImplementation
Program

• Paper form and web-based portal for SPECT-MPI
  data collection, including indications for tests
  and test results
• Analysis of practice patterns based on
  appropriateness criteria
• Feedback of benchmarked practice patterns to
  physicians

39
\
Appropriateness Based on Physician Ordering
40
Relationship between Procedure Indications
Outcomes of PCI ACC/AHA Guidelines
ACC-NCDR
Indications
Anderson et al. Circulation 2005 1122786
41
Relationship between Procedure Indications and
Outcomes of PCI by ACC/AHA Guidelines
ACC-NCDR
Adverse Events
Anderson et al. Circulation 2005 1122786
42
Special Efforts in PCI Outcomes Evaluation
DES/BMS Dual Antiplatelet Therapy

• NCDR Strengths
• Consecutive patients
• Audited data
• Widespread participation gt 1 million/year vs 15k
  clinical trial
• Real life patients (co-morbid conditions,
  older)
• Real life physicians (ask Rob Califf)
• Successful FDA NCDR Groin closure study
• Analytical centers/CV outcomes experts

43
Special Efforts and DES/DAP going Forward

• Missing Elements/Challenges
• Longitudinal Projects/Registries difficult to
  launch
• Patient, Hospital, MD, Industry incentives
• Burden of longitudinal data collection- varying
  models
• HIPAA issues- unique patient identifiers
• IRB approval - not required for In hospital QI
  Registries but would most likely required for
  longitudinal f/u
• Funding, funding, funding, funding
• Registries- good for QI, safety, and measuring
  and benchmarking many outcomes but not
  ideal/challenging for use in clinical trials

44
NCDR Data Merging Partnerships

• AHRQ- DEcIDE Collaborative with DCRI
• NCDR patients
• 600 sites, 2002-2006- 900,000 PCIs of which
  712,000 DES
• Linkage of NCDR with complete Medicare files
• Creating a longitudinal database
• Linkage with HMORN
• Kaiser patient data-pharmacy, costs, and
  longitudinal results
• Real world outcomes assessment tracking
• DES use/outcomes

45
AHRQ- DEcIDE Collaborative with DCRI

• Linkage procedure via probabilistic matching
• Provider , record (unique encrypted
  identifier), DOB, sex, admit/discharge dates
• Match with CMS with very high degree of accuracy
• HIPAA compliant- limited dataset without
  patient direct identifiers (no name or SSN)
• Longitudinal records f/u hospitalizations, death

46
AHRQ- DEcIDE Collaborative with DCRI

• Goals
• Describe temporal trends of DES/BMS
• Analyze downstream DES/BMS patient outcomes
• readmissions, MIs, repeat revascularizations,
  and death
• Role of DAT- length of use post implantation
• Create conceptual model of stent decision making
• Feedback to clinicians-outcomes, workshops,
  publications, education tools, etc

47
AHRQ- DEcIDE Collaborative with DCRI

• Advantages of NCDR large patient base
• Assess low frequency adverse events
• Subgroup patients of interest
• Women
• Minorities
• Diabetes
• Acute coronary syndromes
• Very elderly (gt80years)
• Renal failure