Transcript Improvement In Oncology Practice Performance Through

ASCO
Quality Oncology Practice Initiative (QOPI)
Quality Oncology Practice Initiative (QOPI)
Oncologist-led, practice-based voluntary quality improvement
initiative conceptualized by Dr. Joseph Simone
“Unless one engages practicing physicians in the basic
structure, quality will never become part of the fabric of
practice, the only route to a sustainable quality effort”
Goal: to promote excellence in cancer care by helping oncology
practices create a culture of self-examination and
improvement
(80% of cancer care in the U.S. is provided in community settings)
QOPI: Mission
Develop a means to promote excellence in cancer care
that…
 Is voluntary
 Is designed and run by oncology practitioners





Is relevant and valuable to all practices
Can be delivered anywhere
Measures progress and allows comparison with peers
Is simple and inexpensive to implement
Is not a research project
QOPI Background Information
QOPI measures processes of cancer care in ambulatory
care settings
 Pilot phase: 2002-2005
 National rollout: March 2006
 Three data collections completed
Methodology: semi-annual chart abstraction at the
practice level
Methodology
Practices identify cases of invasive cancer seen within
the last six months
 Sample size based on location FTE’s
 Cases selected as needed to meet minimal
requirements for disease-specific and domain-specific
measures
Practices abstract chart data based on questions
provided either on a standard paper abstraction form or
web-based form
Practices enter de-identified date onsite via a secure
web-based application developed by ASCO
Sample Size
Sample size targets depend on number of medical oncology
FTEs in office (site/location)
 1 med onc FTE:
 2-3 med onc FTEs:
 4-6med onc FTEs:
 7+ med onc FTEs:
48 charts (24 per module)
64 charts (32 per module)
68 charts (34 per module)
80 charts (40 per module)
Measures
Initial measures were selected based on the following
criteria:
 clinically intuitive
 applicable to a large portion of patients
 feasible to evaluate and measure
 accessible by chart abstraction
 amenable to improvement
Measures Sources
ASCO Practice Guidelines and Technology
Assessments
ASCO/NCCN Quality Measures
National Initiative on Cancer Care Quality
NCCN Clinical Practice Guidelines
Consensus-based:
 Literature review
 Organizational priorities (e.g., pain assessment is a
JCAHO standard)
 “Common sense”
Measures by Category
Core measures
 Documentation of care
 Chemotherapy planning and administration
 Pain assessment and control
Disease-specific measures
 Breast cancer management
 Colorectal cancer management
 Non-Hodgkin’s lymphoma management
 Non-small cell lung cancer management
Domain-specific measures
 Care at end of life
 Clinical trial assessment
 Symptom/toxicity management
Core Measures: Examples
Pathology report confirming malignancy available in the
chart
Explicit statement of staging within one month of first
office visit
Documented plan for chemotherapy, including doses and
time intervals
Flow sheet for chemotherapy with doses and blood
counts available in the chart
Some form of patient consent documented
End of Life Measures
Pain assessed on the last or second to last visit before
death
 Pain rated numerically
Patient enrolled in hospice or referred to palliative care
specialist before death
 Patient enrolled in hospice more than 1 week before
death
No chemotherapy administration within the last two
weeks of life
Symptom/Toxicity Management
Serotonin antagonists administered with first
administration of highly emetic chemotherapy
 Corticosteroids added concurrently
Aprepitant administered appropriately with highly emetic
chemotherapy
Prior to administration of erythropoietin or darbepoetin,
documentation of hemoglobin < 11g/dL or anemia/low
hemoglobin symptom
Breast Cancer Measures
Chemotherapy recommended/received for breast cancer
patients less than 70 years of age with tumors >1cm or
axillary lymph node involvement
Trastuzumab recommended for Her2Neu positive breast
cancer patients
Tamoxifen or AI recommended for ER or PR positive
patients
Intravenous bisphosphonates given to breast cancer
patients with bone metastases
Renal function assessed between first and second
administration of bisphosphonates for breast cancer
patients with bone metastases
Colorectal Cancer Measures
CEA measured at least once within 9 months following
curative resection for colon and rectal cancer
Chemotherapy recommended for colon cancer patients
with lymph node involvement
Chemotherapy recommended for rectal cancer patients
with lymph node involvement or penetration through
intestine muscle
Results: Enrollment of Practices
140
ASCO Rollout
120
100
80
60
40
20
0
Pilot 1
Pilot 2
Pilot 3
Open
3/06
Open
9/06
Open
3/07
QOPI: March 2006 Data Collection
87 practices entered data in March 2006
9,324 charts were abstracted
Participants included independent practices, multispecialty groups, academic affiliates, and academic
medical centers
Results: Display Format
Detail
Source
N
QOPI Overall Mean
Practice Median
Box Plot
10%
Mean
Median
Individual
Practices
·
·· ·
10%
50% 80%
Pathology Report Available in Chart
Is there a pathology report
confirming malignancy
available in the chart?
Consensus-based
N= 9357
Mean = 96%
Median = 97%
Explicit Statement of Staging
Is the patient's cancer stage
documented within one month
of his/her first visit to the office
(according to any staging
system or simply the
comment that the cancer is
advanced, metastatic or
incurable)?
Consensus-based
N = 8641
Mean = 89.9%
Median = 92.4%
Documented Plan for Chemotherapy
Is there a plan for the total
amount of chemotherapy to
be given, including doses and
time intervals, which was
documented before the
chemotherapy was started?
Consensus-based
N= 6633
Mean = 76.3%
Median = 87.7%
Some Form of Consent Documented
Is there a signed consent (by
the patient) for treatment in the
chart or a a practitioner's notation
that chemotherapy treatment was
discussed with the patient and
that the patient consented to this
treatment?
Consensus-based
N = 6633
Mean = 87%
Median = 98%
Pain Assessment on Last Visit Prior to Death
Is there a practitioner's
notation documenting the
patient's physical pain or lack
thereof on his/her last visit to
the office before death?
JCAHO
N= 1962
Mean = 74.5%
Median = 75%
Hospice Enrollment: Timing
Among patients enrolled in
hospice, how many were
referred more than one week
prior to death?
Evidence-based
N = 988
Mean = 76.1%
Median = 76.7%
Chemotherapy Administration within the Last
Two Weeks of Life
•
•
•
•
•
Did the patient receive
his/her last dose of
chemotherapy more than
two weeks prior to death?
Evidence-based
N = 480
Mean = 86%
Median = 87%
Use of Serotonin Antagonist Antiemetics
At the first administration of
highly emetogenic chemotherapy , did the patient
receive a serotonin antagonist
type of antiemetic?
ASCO Guideline (evidencebased)
N = 5023
Mean = 97.7%
Median = 100%
Adjuvant Hormonal Therapy for Breast Cancer
Was tamoxifen or an
aromatase inhibitor
recommended for women with
ER + or PR + early stage
breast cancer?
ASCO guideline (evidencebased)
N = 587
Mean = 96.9%
Median = 100%
100%, n=5
Analysis of First Two Data Collections
Round
Participating Centers
Total Charts
abstracted
March 2006
87
9,357
September 2006
113
14,291
Unique practices
participating in both rounds
71
(Study Group)
March
September
7,624
10,240
Study Group
Practice type:
 Independent private practice
54 (77%)
 Academically-affiliated private practice
7 (10%)
 Academic medical center
3 (4.3%)
 Other
6 (8.5%)
Practice size
 Mean MD FTE
7.51 (range 1-34)
 Mean new patients/year
1751 (range 80-11,600)
 Means patients enrolled in
clinical trials
100 (range 0-2300)
All Practices – Mean Performance, All Measures
Spring:
Mean 78.7
Med 80.4
Fall:
Mean 82.3
Med 84.0
(p <0.05 )
Spring
Fall
Improvement
significant in 9
practices by
paired T-test
(None declined
significantly)
(p= 0.010)
Bottom Quartile versus All Others
by Practice
Bottom Quartile
 Number of practices with improvement
 Number of practices with worsening
All Others
 Number of practices with improvement
 Number of practices with worsening
12
0
1
0
(p < 0.05 by paired T-test)
N= 27 measures
*
*
*
*
*
*
*
*
*p < 0.05 by paired T-test
Summary
Voluntary participation in a chart-based practice quality
measurement system is feasible
Disease-based measures show high rates of compliance
Stable performance or continued improvement seen with
serial participation
Practices performing in the bottom quartile demonstrated
the most marked improvement, both in individual
measures and in domains of care
Limitations
Data are self-reported. Verification audits have not yet
been undertaken
Stability of measures over time has not yet been
demonstrated
Relationship of measures to outcome is not definitely
demonstrated
Next Steps
Verification audits
Design and dissemination of oncology practice
enhancement tools
Integration of QOPI into electronic medical records
QOPI: Value Added
ABIM: QOPI participation meets practice
improvement requirement component for recertification (20 points)
CME credit
ACGME: QOPI pilot underway for “practice-based
learning” requirement for fellowship training
Physician recognition status will be awarded by some
insurers for participation