Drug Claims Data and Quality Measurement
Edward Westrick, MD, PhD
MagnaCare Health Services Improvement, Inc
1
Drug Claims Data and Quality Measurement
• Background
– 2 MassPRO special projects
• SCRIPT Project • Dually Enrolled
• Data • Uses in MMA • Recommendations
2
Edward Westrick, MD, PhD
• Internal Medicine • Pharmepi & Pharmecon • HCQIP
– – – – – Quality Partners MassPRO IQH SCRIPT Project AHQA
• MAS X • Leadership-Liaison
• National Panels/Boards
– USP
• Medicare Model Guidelines Committee • Therapeutic Decision Making
– NCQA
• Medication Management Technical Subgroup
– BPS
• Board Cert Pharm
– IOM
• Leading By Example . . .
• PBM • DUR • P&T
Patient Safety: Medication Use
– FMEA – VistA in state hospital system
3
Background: SCRIPT Project
• Blue Bell Meeting1996 – Medication Use in Managed Care – Thought of Prescription Drug Benefit (someday) • “Polypharmacy Meeting” 1997 – Steering Workgroup – Measures Workgroup – Interventions Workgroup • Need for Common Set of Measures Used by All Stakeholders – SCRIPT Project • Test measures – Coalition for Quality in Medication Use • Leverage their use
www.scriptproject.org
4
Coalition for Quality in Medication Use
Member Organizations
• • • • • • • • • • • • • • AMCP AHRQ AAAAI AAFP AAP AACE AAHP AARP ACAAI ACC ACCP ACG ACMQ ACP-ASIM • • • • • • • • • • • • • • ACPM ACR AGA AHQA AHA AMA AMDA AMGA ANA APhA APA ASAM ASA ASCP •ASHSP •ASH •AUA •BCBSA •CDC •CAP •CMS •DOD •DSS •DVA •FDA •HRSA •IHS •JCAHO •MGMA •NACDS •NCCC •NCQA •NCPA •NCL •NPSF •NPC •NQF •OPM •PCMA •USP
5
Coalition Sub-Groups
• • • • • Science Advisors Steering Committee Criteria Workgroup Subcommittee for Models and Testing Subcommittee for Use of Performance Measures • Applications Group • Pharmaceutical Industry Council • Technical Expert Panel
6
SCRIPT Project
• Two Phases
– Measure ID and Selection
• JCAHO
– Measure Testing
• MassPRO
7
Beta Testing
• Ambulatory Setting • Managed Care and FFS • Three Conditions
– CAD; CHF; AFib (post MI, dyslipidemia, hypertension)
8 States
Massachusetts New York Mississippi Louisiana Minnesota Washington Oklahoma West Virginia
8
• Sampling
– 400 cases per condition
• ~5000 charts abstracted
Beta-Blocker Prescribing in Post-Myocardial Infarction: With and Without Exclusions
100 90 80 70 60 50 40 30 20 10 0
91 81 69 62 69 70 76 76 76 82 71 59 63 82 74 81
BB Use without exclusions BB Use with exclusions
B
C
D
E
Collaborator
F
G
H
All
9
Achieving at Least the Target Dose of ACEI
100 All CHF 80
%
LVSD
60
48 46
51 49 49 38 31 31 44 34 29 43 39 44 43 38
40
20 0 A C D
Collaborator
E
F
G
H
All
10
Percent of Follow-Up INR Tests Within 7 Days Following an Out-Of-Range Test (INR > 4.0)
100 80 60
43
% w Follow-up INR within 7 days Mean (31 %)
40
21
29 27 8.8
20 0 A
B
Collaborator
C
E
G
11
End of Project
• Presentation to Coalition in October 2001 was Disrupted by 9/11 • Final Report Feb 2002
– Recommendations
• Reconvene the Coalition for Quality in Medication Use – Present Work of the SCRIPT Project – Discuss Plans for the Future
» » » »
Use of Measures Validation of Measures New Measures Interventions
• Outreach – Presentations at National Meetings – Peer Reviewed Publications – NQMC
12
Special Project in Dually Enrolled
• Demonstrate how merging Medicare and Medicaid data can be useful in quality improvement
– Disparities Studies – Medication Use Studies
Medication Use Studies
Analysis Logic for Administrative Data from SCRIPT Project
13
SCRIPT Measures
Coronary Artery Disease Beta Blockers ACEI Lipid Rx Lipid Test Heart Failure ACEI Beta Blockers Other Meds Monitoring K, RF Atrial Fibrillation Warfarin Monitoring K, RF, INR
14
SCRIPT: Study of Clinically Relevant Indicators for Pharmacologic Therapy: CMS Funded Pilot Project
Medication Use in Heart Failure
Dually Enrolled
100 80 60 40 20 0
E AC
I
Dig r yd H
te tr a /Ni
B AR
o ron i Sp
Di
cs eti ur
BB
Heart Failure: Medication Use
15
ACEI in Heart Failure
DE
100 80
Percent
60 40 20 0
1995
1996
1997
Heart Failure with ACEI Rx
1998
1999
16
Multi Year Group Practice Statewide Comparison: SCRIPT Indicator
17
Warfarin in Atrial Fibrillation
Statewide Multi Year SCRIPT Measures
100
80
Percent
60
DE
40
20
0 1995 1996 1997
AFIB w Warfarin
1998
1999
18
Lipid Lowering Therapy in CAD with Dyslipidemia
Statewide Multi Year SCRIPT Measures 100
80
60
Percent
DE
40
20
0 1995 1996 1997 1998 1999
19
Dyslipidemia w Lipid Lowering Rx: CAD
LDL-C Testing in CAD
Multi Year Statewide SCRIPT Indicator Rates
100
80
Percent
60
DE
40
20
0
1995
1996 1997 1998 CAD w Cholesterol Test
1999
20
Group Practice Statewide Comparison: SCRIPT
21
Data Needed to Compose Measures
• Diagnoses and Procedures from Parts A & B • Pharmacy Claims from Medicaid
– – – – – – – Medication Name Dispensing Date Strength Days Supply Quantity Dispensed Prescriber Pharmacy
Will need to have equivalent elements from MMA implementation
22
Recommendations
• Develop Measurement System for Multiple Levels of Analysis Using Claims Data
– – – – –
– – – –
National State Practice Plan Year
Criterion Validity – Chart Abstraction Construct Validity – Longitudinal Claims-based Analyses Content Validity – Expert Panels, Lit Reviews Face Validity – Expert Panels
• Validate the Claims-based Measures
• Develop System Using Dually Enrolled Population, in Preparation for Eventual Use in MMA
23
Use of Measures at Different Levels of Analysis
• National
– CMS, Congress, NHQR
• ? Plan
– Accountability
• State
– QIO Evaluation in Future SoWs – DUR (Medicaid)
• ?? Patient
– Assist in Clinical Intervention
• Registries, Reminders
• Practice
– Select Collaborators for Improvement Work – Performance Feedback
• ??? CSA
• Year
– Run Charts
24
Contact Info
Edward Westrick, MD, PhD MagnaCare HSI, Inc. ewestrick@pol.net 401-475-3734 www.scriptproject.org
25