* This presentation is prepared by the author in one’s personal capacity
for the purpose of academic exchange and does not represent the views of
his/her organisations on the topic discussed.
Background
Shortfall across a broad range of measures
Paid by capitation- quality improvement activities not
billable
Benchmarking – performance is low; variations in
adherence to recommended treatment patterns
Çpublic release of informationÆ but patients fail to use
information to vote with their feet
Çhealth plans in the US adopted pay-for performance
mechanisms
Early Experience with Pay-for –
Performance
From Concept to Practice
Rosenthal MB, Frand RG, Li Z, Epstein AM
JAMA 2005; 294: 1788-1793
Dr W K CHING
1
Dr W K CHING
What is already known?
Question
Limited evidence
RCT – rare with a few demonstrating that pay for
performance ÆÇ quality
Difficult to enrol large no of paying organisations
RCT, if any, focus on single indicator/ aspect of care
(versus multiple conditions in real life)
Dr W K CHING
Dr W K CHING
4
Pay-for-Performance –
New Quality Incentive Program
One of health plans in the Integrated Healthcare Association
initiative
Share information since 1995
Since 2003- bonus of $2/ month per PCC member -
Certain
Number
of
members
Cervical cancer screening
Mammography
Childhood immunisations
Hb A1C for DM
LDL cholesterol for coronary artery disease
Satisfaction with medical group
Satisfaction with primary care physician (PCP)
Satisfaction with referral process
Satisfaction with specialist
Effective PCP communication
Dr W K CHING
Reward providers according to :
• attainment of a predetermined level of
performance or
• improvement?
3
PacifiCare of California (PCC) Quality
Incentive Program
–
–
–
–
–
–
–
–
–
–
2
10
quality
measure
s
Meet 75th
percentile of
2002
performanc
e
8
9
No bonus
$0.23/ member
per month
Up to
5
Dr W K CHING
5% of professional capitation
0.8% of overall revenue 6
1
Comparison
QIP announced
for California
Apr 01
California
Network
Evaluation
QIP started for
California
Jan 02
Pre-QIP
Jan 03 Apr 03 Oct03
Post-QIP
10 among quality
measures
Changes in
3 quality
measures
Individuals who
should receive
Mammography
Changes in
3 quality
measures
Pacific
Northwest
Network
=
Individuals who
did receive
HbA1C
vs
PacifiCare
Cervical
cancer
screening
First year –end evaluation
In training examination
scores
Dr W K CHING
7
Dr W K CHING
8
Improvement in Clinical Quality Scores for Quality Incentive Program (QIP) Measures
Pre-QIP, %
Post-QIP, %
Difference
(Post – Pre), % (SE)
P
Value
<.001
Cervical cancer screening
RESULTS
California (n = 134)
39.2
44.5
5.3 (1.6)
Pacific Northwest (n = 33)
55.4
57.1
1.7 (0.9)
.03
Difference
-16.2
-12.6
3.6 (1.8)
.02
.04
Mammography
California (n = 134)
66.1
68.0
1.9 (1.1)
Pacific Northwest (n = 32)
72.4
72.6
0.2 (1.1)
.43
Difference
-6.3
-4.6
1.7 (1.5)
.13
.02
Hemoglobin A1C testing
Dr W K CHING
California (n = 134)
62.0
64.1
2.1 (1.0)
Pacific Northwest (n = 31)
80.0
82.1
2.1 (3.3)
.20
Difference
-18.0
-18.0
0.0 (3.5)
.50
9
Dr W K CHING
10
Level of Baseline Performance
Financial Awards Offered and Made in the First Year of the Quality Incentive Program (QIP)
Maximum possible
network award, $
Total amount paid, $
Mean group payment
Maximum group payment
No. of eligible groups
No. of groups to reach any
medical group quality target
No. of groups to reach ≥ 5
medical group quality targets
July
2003
October
2003
January
2004
April
2004
Total QIP
Year 1
3 251 092
3 334 328
3 379 169
3 253 095
12 850 505
812 772
833 582
844 792
935 079
3 426 226
4986
5083
5183
5437
20 702
62 767
73 129
72 572
86 872
305 702
163
164
163
172
172
97
110
118
129
129
13
15
14
14
15
Dr W K CHING
11
At or above
target
All groups
Below target
but not more
than 10%
>10% below
target
High
Group 1
Middle
Group 2
Low
Group 3
Dr W K CHING
12
2
Results
Workload
High
Middle
Low
Group 1
Group 2
Group 3
75%
20%
5%
Quality Improvement After the Quality Incentive Program (QIP) and Bonus Payments to California Groups
With High, Middle or Low Baseline Performance
Pre-QIP
Rate, %
Post-QIP,
Rate %
Improvemen
t (Post-Pre),
% (SE)
P
Value
Bonuses Paid in
Year 1, $
Group 1 (n = 50)
597 091
53.6
56.0
2.5 (0.8)
.001
436 618
Group 2 (n = 32)
287 610
40.8
48.1
7.4 (2.4)
.001
127 632
Group 3 (n = 52)
305 041
23.0
34.1
11.1 (3.9)
.002
26 859
Group 1 (n = 43)
557 119
72.3
73.0
0.7 (0.9)
.22
383 370
Group 2 (n = 50)
384 852
64.9
67.2
2.3 (1.0)
.01
88 787
Group 3 (n = 40)
244 270
52.6
59.1
6.6 (4.1)
.05
987
Group 1 (n = 46)
547 687
75.4
77.1
1.8 (1.2)
.07
360 155
Group 2 (n = 26)
231 157
62.2
64.8
2.7 (2.3)
.12
101 619
Group 3 (n = 56)
395 450
39.4
49.2
9.8 (2.7)
<.001
53 218
Quality Domain
Cervical cancer screening
50%
25%
Mammography
25%
Hemoglobin A1C testing
Improvement in performance – greatest for those with low
baseline performance
Incentives – directly proportional to baseline performance
Dr W K CHING
Total
PacifiCare
Members
Bonus
13
Dr W K CHING
14
Discussions
• Confounders? Per-QIP trend of both networks
similar
• Infrastructure & staff leading to improvement
takes time to be acquired
• Incentive too low
• Ceiling effects of those good performer
• Uniqueness of PacifiCare- generalisable?
• Hence the incentive design is the key to success
Dr W K CHING
15
3