1. How Performance Data Can
Influence Clinical Behavior
Kirk Stapleton- SVP United HealthGroup
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

2. Unexplained Market Variation in Cost and Quality
2
Medicare Per Enrollee Annual Cost Growth Rates Nearly 300% Variation in Last 2 Years of
+3.7% to 7.7% Life by Market
Miami $81,175 per Person
Minneapolis $33,325
Honolulu $27,655 per year
Health Expenditure Data, Health Expenditures by State of Residence, Centers for Medicare and Medicaid Services, Office of the
Actuary, National Health Statistics Group, released September 2007; available www.cms.hhs.gov/NationalHealthExpendData/“
Kaiser Foundation, http://www.statehealthfacts.org/profileind.jsp?ind=332&cat=6&rgn=25
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

3. Review of Programs
3
• Organ Transplant- Hospital & Clinical Team Focus
• Cardiac Data Sharing- Office Practice Focus
• Cardiologist Gainsharing on AMI- Practice vs Hospital Gain
• Provider Designation- Patient Focus
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

4. Transplant Centers of Excellence (since 1986)
4
Goal:
•Provide payors and patients with accessible, high quality
performance with economic value and consistency for S upe rior Ou tc om e s, C linica l Ex pe rtis e a nd Exp erie nce
transplantation services le ad to Gr ea te r S av ings a nd C os t A voida nc e
 C en te rs o f Ex ce ll e nc e N e t w o rk p ro g ra m s y ie ld a n a v e ra ge 2 2 %
d e c re as e in ho s p i tal le n g th o f s ta y
Specialized Transplant Care Provider Network  C li n ic a l Ex p e rt is e l e a ds to a 21 % re d u c tio n o f inc i d e n c e b y
a vo id i n g u nn e ce s s a ry a n d i n a p p ro pr i ate tra n sp la n ts, thr ou g h b e tt er
•126 Multi-organ Medical Centers (623 transplant programs) d ia g no s is a n d m o re ap p ro p ria te tre atm e n t i de n tif ic a ti o n
•selected based meeting quality criteria (ie. patient and graft Tr a n s pla n t I n p a tie n t D a y s Tr a n s pl a nt In c id e nc e pe r 1 M M M e m b e r s
survival thresholds, case volumes, team attributes), and; 25
180
• economic criteria (patient episode of care contracts including 20
22%
160
21%
Txps per 1M M Memb ers
140
Avg. Txp. IP Days
15 120
MD, organ procurement, inpatient and outpatient care)- 10
100
80
•Patient episode of care contract with Medical center~ pre- 5
60
40
20
evaluation through 1 year followup 0
O pt u m He a lth CO Es M i lliim a n 20 0 5 Es ti m a te
0
O p t um He a lt h M illim a n 200 7
In ci de n c e Es t im a t e
Patient and Referring MD Decision Support S OU R CE S : 2 00 2 – 2 00 7 Op tu mH ealt h C O E tra ns pla nt da ta , M il lim an 2 00 5 + 2 00 7
13
•Patient education and MD referral support to transplant
network with dedicated case and account management and Savings via Contract Design Expertise
service
Claims Repackaging OptumHealth’s transplant cost exposure experience allows us to design
•fee for service claims assembled into patient episode and detailed terms around other cost risk areas outside of the transplant
procedure which often represent 40%+ of costs
priced consistent with contract terms- reported and measured
e.g. Fixed $ Defined Cont rol led Cost s Uncapped Cost Risk
against market and billed charge costs.
e .g. % Disc ount
Phase I Phase II Phase III Phase IV Phase V
Transplant Pre- Transplant Post- Transpl ant Post-Transplant
T o t a l R e f e r r a l s a n d T r a n s p la n t s 2 0 0 1 - 2 0 0 7 Evaluat ion Transplant Procedure 90 Days 365 Days
Dis tr ib uti on o f
1 2000
R e f e r r a ls
1 1 ,1 4 0
Tra ns p la nt
Cha rg e s 5% 10% <60% 15% 10%
10, 686
T ra n s p la n t s
976 2 ** ** Average
Average
1 0000 OH **
savings per
savings per
867 5 Contracts
T o ta l p e r Y e a r
case* ==43%
case* 43%
8000
Competitor Average
647 0 Average
Con tracts savi ngs per
5 738 savings per
6000 522 2 case* ==30%
* Sav in gs based on Milliman, In c charges (2006), which do not incl. candidacy ch arges case* 30%
** For in patient services; outpatient is % discounts a n also uncapped
d
368 8 3, 960
36 27
4000
30 68 Typical Competitor Contracts address <60% of the ec onomic risk
2 542
19 55 21 55
2000
16
0
0 1 02 03 0 4 0 5 06 0 7
20 20 20 2 0 2 0 20 2 0
Confidential property of UnitedHealth Group.Y e a r distribute or reproduce without the express permission of UnitedHealth Group.
Do not

5. Cardiac Data-Sharing Visit Program (since 2006)
5
Goal:
•decrease utilization of unnecessary high-cost diagnostic services and
procedures Results:
•redirect inpatient procedures to low-cost facilities •A paired t-test analysis showed a statistically
•reinforce society defined appropriateness criteria significant difference in the change pre/post for
Angiograms, Echo's, and Perfusion Studies per
•Activity: Office Visit for the intervention group compared to
• Study: 2 groups (visited /35 groups~351MDs vs not visited/16 the control group No difference in use of low cost
groups~200MDs ) facilities and cv rate per visit noted
•Group visits discussions included ,episode of care measurement, utilization
of diagnostic services, performance characteristics and comparison to peers, •In addition, the intervention group showed a
ACC treatment guidelines statistically significant decrease in PCI
Procedures/Office Visit
•Intervention vs control groups compared on pre/post-intervention utilization
metrics •Overall rate of use per office visit declined 16% for
Diagnostic Services/Office Visit*: Angiograms, Echocardiograms, Perfusion Studies angiograms and 6% perfusion studies for the
Procedures/Office Visit: CV Surgeries, PCIs intervention group, offset by an increase of 13% of
% of Procedures performed at a Low Cost Facility: CABG, Valves, Implants, PCIs echoes. This trend appears sustainable at 12
months..
•12-18 month group and individual MD risk adjusted claims data , grouped
into patients episodes; groups are selected based on variance between •Program expanded to over 400 groups nationwide
actual and expected episodic costs as measured in the Premium program
Cardiac Data-Sharing Preliminary Results - (Paired T-Test, Two-Tailed)
Pre/Post Visit Change
Metric Pilot Group Control Group Fav/Unfav Significance
% of Cases Performed at a "Low Cost" Facility (met efficiency) n/a n/a Not applicable T-Test not significant for both
Diagnostic Services
Angiograms / Office Visit + 11 bp n/a Favorable T-Test significant for Pilot only
Echocardiograms / Office Visit - 27 bp - 37 bp Favorable T-Test significant for both
Perfusion Studies / Office Visit + 26 bp n/a Favorable T-Test significant for Pilot only
Procedures
CV Surgeries / Office Visit n/a n/a Not applicable T-Test not significant for both
PCIs / Office Visit + 5 bp n/a Favorable T-Test significant for Pilot only
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

6. Cardiac Gainsharing Pilot- Tampa (since Jan 2009)
6
Goal:
•Improve quality and cost effectiveness of cardiac care with support of community delivery systems (cardiologist) through
sharing cost and quality performance data
•Participating groups must meet threshold of quality and demonstrate consistency of high quality
•Participating groups share in market level savings through enhanced fee for service
Establish mandatory quality thresholds for all participating cardiologists
•Must meet 80% of cardiac related Evidence Based Medicine criteria
•Must earn UHPD Premium Designation for Quality for all proceduralists
Created a mutually shared target for total cardiac cost for all members living in the geographic boundary
•Cardiac PMPM target shared by key cardiologist physician groups in the market
•Includes inpatient facility costs, all professional fees (regardless of specialty), ancillary testing, etc.
•Provide opportunity for physicians to increase revenue while helping improve overall quality and cost for members and employers
•Group of MDs & UHC are focused on reduction of Chest Pain Admissions
Gain Sharing Bonus Determination Early but Promising reductions in Cardiac Admissions
The Gain Sharing bonus will be determined by the 6 month Cardiac PMPM cost for Pinellas County Weekly Chest Pain Admissions
Pinellas county members. There is no penalty or “down-side” to the physician groups.
y = -0.0427x + 1712.2
Baseline Cardiac 20
PMPM Costs for If actual costs finish above
Market target; no bonus payment
18
Chest Pain Admissions 16
Target PMPM Costs
14
Costs below target;
12
Participating MD Groups get 50% of the
savings below the target level 10
Example: if target was $30.00 and actual costs
for the 6 month measurement period were 8
$28.00 PMPM, the bonus for physicians would
be 50% X (30.00 – 28.00) = $1.00 PMPM* 6
(~$1.4 million for pilot population)
4
2
0
*The bonus is distributed through a temporary increase in the physician’s fee schedule
9/1/2008
9/15/2008
9/29/2008
10/13/2008
10/27/2008
11/10/2008
11/24/2008
12/8/2008
12/22/2008
1/5/2009
1/19/2009
2/2/2009
2/16/2009
3/2/2009
7
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

7. Cardiac and Orthopedics Specialty Centers (Since 2004)
7
Goal:
•Differentiate hospital/MDs network for acute and frequent interventions resulting in value creation for customer and health consumer
(UHC)
•Share data with delivery system to Improve practice patterns through public designation supported with actionable information –
gain cost efficiency through reduced complications, and rework
Facilities and MDs that are measured against national When combined with Premium
performance quality standards and cost expectations: Designations for MDs who achieved quality
and efficiency and high performing cardiac
•Facility and Clinical teams measured together- scored and depicted on
and orthopedic facilities :
quality scale from 1 to 3 stars; economic scores depicted from greater
efficiency to lower efficiency Cardiologists who earn a quality designation have 42% fewer
redo procedures and 13% lower complication rates for stent
placement than other cardiologists.
•Cardiac programs (1,240 interventional, rhythm management and surgical)
nationwide that are scored against criteria utilizing, process and outcomes Cardiothoracic surgeons who earn a quality designation
have 19% fewer redo CABG (CoronaryArtery Bypass Graft)
measures ~ Society for Thoracic Surgery ( 11 NQF measures) and surgeries than other surgeons.
American College of Cardiology (7 executive summary measures) hospital
ranking to measure quality and outcomes; uses UHC claims data for risk Cardiologists who earn a quality designation have 31% fewer
redo procedures for rhythm management device implants
adjusted episode of care costs (pacemakers, implantable defibrillators, etc.) than other
cardiologists
•Spine and total joint repair programs (523) uses process of care and Orthopedic surgeons who earn a quality designation have
outcomes (complication, redo) and volume data survey data ; and UHC 45% fewer redo spine surgeries than other surgeons
claims data for risk adjusted cost
Orthopedic surgeons who earn a quality designation have
20% fewer redo arthroscopic knee procedures than other
surgeons.
•Hospitals with the highest quality ranking and lowest costs
are projected to have hospital costs that are 30% below market
average
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.

8. What We Have Learned
8
• Reporting Accuracy- Patient Level Proof
• Reconsideration- If Data Is Wrong- Get it Right
• Clinical Reporting Is A Process
 Introduction to Process- “What we are going to measure & Why”
 Transparency of methodology- “How we are measuring and When”
 Face to Face Meeting with Actionable Information “Your Report and What it Means”
• Academic and Health Policy Recognition
• Reinforcement for Sustainable Change –New Business Model Adoption
 Policy and Payment
 Market Share (ie More Patients)
 Recognized Value (ie improved profitability or greater revenues)
Confidential property of UnitedHealth Group. Do not distribute or reproduce without the express permission of UnitedHealth Group.