1

Hospital P4P: The CMS/Premier Hospital Quality Incentive

Demonstration Project

March 10, 2009

Mary B. BergersonRegional Quality DirectorSt. Helena Hospital

2

Nationwide knowledge to improve local healthcare

2,000 Hospitals gain the advantages of national scale by uniting through the Premier healthcare alliance

3

2007

Value-based Purchasing timeline

1990’s 2002 20072003 2004 2005 2006

Medicare Modernization Act ties hospital market basket updates to quality reporting

Institute of Medicine (IOM) ReportTo Err is Human CMS

Report to Congress on VBP

2008

Premier Hospital Quality Incentive Demo launched

Hospital Comparelaunched

Deficit Reduction Act mandates a Report to Congress on the Plan to Implement a Medicare Hospital VBP Program

Grassley-Baucus (S.1356) value-based purchasing legislation introduced

Senate Finance Committee Roundtable on VBP

2009

MedPACReport supports use of VBP IOM

Report on Aligning Hospitalincentives

4

• CMS and Premier partnership project

• First national Pay-for-Performance (P4P) demonstration

• Tests the hypothesis that financial incentives and public recognition can increase quality of care

• A three-year effort launched October, 2003

• Approximately 260 hospitals in 38 states

CMS/PremierHospital Quality Incentive Demonstration

5

Rewarding delivery of widely accepted evidence-based clinical indicators

Acute myocardial infarction (AMI)1. Aspirin at arrival2. Aspirin prescribed at discharge3. ACEI/ARB for LVSD4. Smoking cessation advice/counseling5. Beta blocker prescribed at discharge6. Beta blocker at arrival7. Thrombolytic received within 30 minutes of hospital arrival8. PCI received within 90 minutes of hospital arrival9. Inpatient mortality rate

Coronary artery bypass graft (CABG)1. Aspirin prescribed at discharge2. CABG using internal mammary artery (Test)3. Prophylactic antibiotic received within one hour prior to surgical incision4. Prophylactic antibiotic selection for surgical patients5. Prophylactic antibiotics discontinued within 24/48 hours after surgery

end time6. Inpatient mortality rate7. Post operative hemorrhage or hematoma8. Post operative physiologic and metabolic derangement

Heart failure (HF)1. Left Ventricular Systolic (LVS) assessment2. Detailed discharge instructions3. ACEI or ARB for LVSD4. Smoking cessation advice/counseling

Pneumonia (PN)1. Percentage of patients who received an oxygenation assessment

within 24 hours prior to or after hospital arrival2. Initial antibiotic selection for Community Acquired Pneumonia3. Blood culture collected prior to first antibiotic administration4. Influenza screening/vaccination5. Pneumococcal screening/vaccination6. Antibiotic timing, percentage of pneumonia patients who received

first dose of antibiotics within four hours after hospital arrival7. Smoking cessation advice/counseling

Hip and knee replacement1. Prophylactic antibiotic received within one hour prior to surgical

incision2. Prophylactic antibiotic selection for surgical patients3. Prophylactic antibiotics discontinued within 24 hours after surgery

end time4. Post operative hemorrhage or hematoma5. Post operative physiologic and metabolic derangement6. Readmission within 30 days to any acute care facility

Surgical

Italics = outcomes measure

6

Dramatic and Sustained Improvement

HQ

ID C

ompo

site

Qua

lity

Scor

e

Avg. improvement across all 5 clinical

areas for median CQS (19 quarters)

18.66% Clinical Area

Improvement (percentage points)

AMI 8.9%

CABG 14.1%

Pneumonia 25.9%

Heart Failure 31.4%

Hip & Knee 13.0%

CMS HQID Composite Quality Score

CMS/Premier HQID Project Participants Composite Quality Score: Trend of Quarterly Median (5th Decile) by Clinical Focus Area

October 1, 2003 - June 30, 2008 (Years 1, 2, & 3 Final Data; Years 4 and 5 Preliminary Data)

89.6

%

85.1

%

70.0

%

64.0

%

85.1

%

90.0

%

85.9

%

73.1

%

68.1

%

86.7

%

91.5

%

89.4

%

78.1

%

73.1

%

88.7

%

92.5

%

90.6

%

80.0

%

76.1

%

90.9

%93.5

%

93.7

%

82.5

%

78.2

%

91.6

%

93.4

%

94.9

%

82.7

%

81.6

%

93.4

%

95.1

%

96.2

%

84.8

%

83.0

%

95.2

%

95.7

7%

97.0

1%

86.3

0%

84.3

8%

95.9

2%

96.0

%

96.8

%

88.5

%

86.7

%

96.6

%

96.1

% 98.3

%

89.3

%

88.8

%

97.1

%

96.8

%

98.4

%

90.1

%

90.0

%

97.8

%

96.8

%

98.4

%

91.4

%

89.9

%

97.9

%

97% 98

%

92%

90%

98%

97.0

%

97.7

%

92.4

%

91.6

%

97.9

%

97.6

%

97.8

%

93.5

%

93.2

%

98.0

%

97.5

%

98.4

%

93.4

%

93.4

%

98.1

%

98.3

%

98.5

%

94.2

%

94.2

% 97.4

%

92.3

%

98.2

7%

99.0

1%

94.8

5%

94.9

0% 97.4

6%

94.1

1%

98.5

4%

99.1

9%

95.9

0%

95.3

8% 98.1

6%

95.2

7%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100%

AMI CABG Pneumonia Heart Failure Hip and Knee SCIP

Clinical Focus Area4Q03 1Q04 2Q04 3Q04 4Q04 1Q05 2Q05 3Q05 4Q05 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07

1Q08 2Q08

7

More Patients are Receiving Every quality measure

Evidence-based Care ImprovementsAvg. improvement from

4Q03 to 2Q08 in all clinical areas(19 quarters)

55.05%Clinical Area

Improvement(percentage points)

AMI 23.7%

CABG 66.5%

Pneumonia 65.1%

Heart Failure 54.9%

Hip & Knee 65.1%

App

ropr

iate

Car

e Sc

ore

CMS/Premier HQID Project Participants Appropriate Care Score: Trend of Quarterly Median (5th Decile) by Clinical Focus Area

October 1, 2003 - June 30, 2008 (Year 1, 2, and 3 Final Data; Year 4 and 5 Preliminary)

70.7

%

30.0

%

22.3

%

34.7

%

27.8

%

72.7

%

34.1

%

28.0

%

43.6

%

34.1

%

75.7

%

45.8

%

34.7

%

50.0

%

41.2

%

80.0

%

48.7

%

39.0

%

53.8

%

53.6

%

80.9

%

68.5

%

43.8

%

58.5

% 63.6

%

80.6

%

77.3

%

44.3

%

62.6

%

72.1

%

85.0

%

82.9

%

50.7

%

64.6

%

78.6

%87.0

%

84.2

%

53.8

%

68.0

%

81.3

%87.8

%

86.6

%

60.9

%

72.3

%

85.7

%

88.2

%

91.9

%

62.8

%

75.8

%

85.9

%

89.6

%

93.3

%

67.6

%

78.1

%

89.5

%

88.6

%

91.7

%

70.3

% 78.3

%

87.1

%

90.0

%

91.7

%

82.6

%

79.2

%

90.0

%

90.0

%

93.3

%

82.8

%

82.5

%

86.4

%92.1

%

94.1

%

87.0

%

85.2

%

86.2

%92.8

%

95.5

%

87.0

%

86.0

%

87.0

%93.5

%

92.7

%

77.1

%

85.5

%

86.9

%

76.7

%

93.8

%

96.0

%

82.7

% 87.9

%

89.3

%

84.0

%

94.4

%

96.5

%

87.4

%

89.7

%

92.9

%

84.0

%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

AMI CABG PN HF Hip and Knee SCIP

Clinical Focus Area

4Q03 1Q04 2Q04 3Q04 4Q04 1Q05 2Q05 3Q05 4Q05 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07

1Q08 2Q08

8

Substantial Improvement by Lower Performers

Percentage of Hospital Clinical Focus Groups in "Penalty Box"

HQID Final Data Years 1-3

20%

11%8%

5% 5%3% 2% <1%1%

0%

5%

10%

15%

20%

25%

Year 1Base

Year 2Q1

Year 2Q2

Year 2Q3

Year 2Q4

Year 3Q1

Year 3Q2

Year 3Q3

Year 3Q4

Quarter

Perc

enta

ge

Percentage of Hospital Clinical Focus Groups in "Penalty Box"Preliminary HQID Year 5 Results

Updated 2-4-09

4.4%

3.3%3.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

Q4-07 Q1-08 Q2-08 Q3-08

Quarter

Perc

enta

ge

Percentage of Hospital Clinical Focus Groups in "Penalty Box"Preliminary HQID Year 4 Results

Updated 2-4-09

2.4%

1.6%

1.1%

1.4%

0.8%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

Q4-06 Q1-07 Q2-07 Q3-07 Yr4 Prelim

Quarter

Perc

enta

ge

9

A composite of 19 measures shared in common between HQID and Hospital Compare shows P4P hospitals performing above the nation as a whole

In Broader Comparison, HQID Hospitals Excel

National Leaders in Quality Performance

• HQID participants avg. 6.8% higher than Non-Participants

• Avg. improvement for HQID participants = 9.7%

• Avg. improvement for Non-participants = 7.4%

New England Journal of Medicine publication by Lindenauer et al. (February 2007) found that hospitals engaged in P4P achieved quality scores 2.6 to 4.1 percentage points above other hospitals due solely to the impact of P4P incentives.

HQID hospitals have higher quality ratings than national hospitals overall

10

• Established in 1878, St. Helena serves a five county region and 200,000 residents in largely rural northern CA

• 158 inpatient beds treating 75,000 inpatient & outpatient visits annually

• Medical staff of 125 with 920 total employees

• 63% revenue from Medicare and Medicaid

11

Continuous Improvement in Composite Quality Scores

Continuous improvement from Year 1 to Year 5 Year to Date (YTD) HQID data for the Composite Quality Score (CQS), a combination of clinical quality measures and outcome measures.

St Helena HospitalTrend of Quarterly HQID Composite Quality Scores by Clinical Focus Area

October 1, 2003 - June 30, 2008 (Year 1-3 Final Data; Year 4 & 5 Preliminary Data)

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100%

Com

posi

te Q

ualit

y Sc

ore

AMI Heart Failure Pneumonia CABG Hip/Knee

▲ = HQID Top 20% Threshold Value; ● = HQID Median

12

Consistent Top Performer

• Top Performer in Heart Failure for Years 1 and 2• Top Performer in AMI in Years 2 and 3

– Tracking for Top Performer in Years 4 and 5

• Top Performer in Hip/Knee Replacement for Year 3– Tracking for Top Performer in Year 4

• Tracking Top Performer in CABG for Year 5• Significant improvement in Pneumonia

– 65% in Year 1 to 94% Year 4– Tracking for Top Improver Award in Years 4 and 5

• Tracking to receive Attainment Award for all clinical areas in Year 4 and all except Hip/Knee Replacement for Year 5

13

More Patients are Reliably Receiving Evidenced-based Care

The appropriate care score (ACS), also referred to as “perfect process or “all or nothing” to designate when a patient receives all possible care measures within a clinical area, showed improvement across time.

St Helena HospitalTrend of Quarterly HQID Appropriate Care Scores by Clinical Focus Area

October 1, 2003 - June 30, 2008 (Year 1-3 Final Data; Year 4 & 5 Preliminary Data)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

App

ropr

iate

Car

e Sc

ore

AMI Heart Failure Pneumonia CABG Hip/Knee

▲ = HQID Top 20% Threshold Value; ● = HQID Median

14

Performing Above the National Average

A composite of 19 process measures shared between HQID and Hospital Compare shows St. Helena Hospital performing above the nation as a whole.

St. Helena Hospital Compared to National Group TrendHospital Compare Data

19 Process Measures Aggregated to Overall Composite Process Score86

.7%

87.5

%

88.3

% 89.7

%

90.3

%

83.0

%

84.0

%

85.1

% 86.4

% 88.1

%

89.1

%90.6

% 92.5

%

93.0

%

94.2

%

94.7

%

85.9

%

91.7

%

76%

78%

80%

82%

84%

86%

88%

90%

92%

94%

96%

July 05-June 06 Oct 05-Sept 06* Jan 06-Dec 06 Apr 06-Mar 07 July 06-June 07 Oct 06-Sept 07

Comparison of HQID Participation and Non-Premier Status

Com

posi

te P

roce

ss S

core

(Mea

n Va

lue,

Per

cent

)

National Average State Average St Helena Hospital

*Beginning w ith Oct 05-Sept 06 the influenza vaccination measure became unsuppressed and the number of process measures increased from 18 to 19

15

Improvement and Savings

Avg. cost improvement across all clinical areas

$1,063

If all hospitals in the nation were to achieve this improvement, the estimated cost savings would be greater than $4.5 billion annually with estimated 70,000 lives saved per year

Avg. improvement in mortality across four clinical areas

1.87%Clinical Area Improvement

AMI $1,599

CABG $1,579

Pneumonia $811

Heart Failure $1,181

Hip Replacement $744

Knee Replacement $463

Clinical Area ImprovementAMI 2.27%

CABG 0.95%

Pneumonia 2.39%

Heart Failure 1.86%

16

Findings: Mortality, Complications, Length of Stay and Costs all go down for Heart Attack (AMI)

12000

12500

13000

13500

14000

14500

15000

15500

Q4-03 Q1-04 Q2-04 Q3-04 Q4-04 Q1-05 Q2-05 Q3-05 Q4-05 Q1-06 Q2-06 Q3-06

Median Cost per Case over 3 years (AMI)

17

Conclusions

1. Creates a performance improvement engine• Public reporting• Financial incentives

2. Aligns incentives within hospitals 3. Re-aligns payment incentives in Medicare

• From rewarding more procedures to rewarding quality procedures

4. Improved quality is associated with saving lives and reducing costs

18

Policy Recommendations

1. Create a positive incentive to improve performance. – All unallocated funds must be used to reward top performers and improvers– There must be an annual actuarial assessment to identify savings, which should be

used to fund a bonus pool– Phase in so that hospitals can realistically achieve the benchmarks

2. Align physician and hospital interests. – Assure alignment between physician and hospital measures– Hospitals should be able to share money from the bonus pool with their physicians

3. Set benchmarks based on real world evidence from the CMS/Premier HQID project.

4. VBP should be irrevocably tied to public reporting. – The Hospital Compare Web site must be more user friendly– Hospital Compare should include reporting of the hospital’s performance in

delivering all recommended quality measures for each clinical condition– All new measures should be tested and publicly reported use in a VBP program

5.  Government should direct attention and resources to lower performing hospitals

– QIOs should be directed to focus attention on non-performing hospitals