PROMETHEUS Pilot Pre- conference · Proprietary & Confidential. Health Care Incentives Improvement Institute, Inc. PROMETHEUS Pilot Pre-conference March 23rd 2011

Proprietary & Confidential. Health Care Incentives Improvement Institute, Inc.

PROMETHEUS Pilot Pre-conference

March 23rd 2011


Agenda

Session Topics/Title Presenters

8:30-8:45a.m. ¨  Five lessons learned from the RWJF-funded Prometheus Payment pilots

•  Francois de Brantes(HCI3)

8:45-9:15a.m. ¨  Analyzing opportunities for episode of care (EOC) payment

•  Elizabeth Siteman (HCI3)

9:15-10:15a.m. ¨  Contracting for episodes •  Doug Emery (HCI3) •  Eddie Bell (Providence) •  Greger Vigen (SOA)

10:15-10:30a.m. ¨  Break

10:30-11:30a.m. ¨  Operationalizing & Implementing EOC payment

•  Chad Brown (HCI3) •  Liz Rubin (Horizon) •  Joe O’Connor (CareFuze)

11:30a.m.-Noon ¨  Q&A •  All

2


Lessons learned from Implementations •  5 ingredients to success:

– Full CEO engagement – Commitment by willing plan AND provider – Clean and complete claims and eligibility data – EMR systems – Sense of urgency

•  We’ve actually made the operations relatively easy, and they’re becoming easier.

•  However, hospitals, in particular, have a growing dilemma to manage

3


The Hospital CFO’s Dilemma

•  It’s better to make a 7% margin on $140 million in revenue than a 4.5% margin on $190 million in revenue

•  However, you’re worse off if you’re making a 7% margin on $40 million plus a 4.5% margin on $120 million

•  So…how do you manage the transition from volume-based payment to value-based payment?

4


VBP volume and margin dictate the sustainable pace of change

$0

$20,000,000

$40,000,000

$60,000,000

$80,000,000

$100,000,000

$120,000,000

$140,000,000

$160,000,000

$180,000,000

$200,000,000

1 2 3 4 5 6 7 8 9 10 11 12 13

Total Revenue

Break-even FFS Volume

EOC Volume

Total Margin

Note: Assumes a 7% margin for Episode of Care payment and a 4.5% margin for FFS payments. FFS b/e Vol = (Target $ Margin – (EOC $ Vol * EOC margin)) / FFS margin.

5


For contact information: www.HCI3.org www.bridgestoexcellence.org www.prometheuspayment.org


Analyzing Opportunities for Episode of Care Payment

National P4P Summit Preconference II March 23, 2010


Agenda

•  Identifying Sources of Variation in Episode Costs

•  Analyzing Potentially Avoidable Complications

•  Provider Level Analysis as a guide to contracting, performance assessment, and quality improvement


Sources of Variation in Episode Costs •  Provider Performance in terms of potentially

avoidable complications (PACs) •  Unit Price per Service – determined by fee

schedules or contracted rates •  Severity of the patient – sicker patients may

require more resources for appropriate care •  Provider Practice Patterns in terms of

frequency and types of services provided for routine or typical care

3


Variation in COPD Episode Costs across 10 Provider Groups

4

Average annual episode costs per patient are a routine measure used to assess cost performance across providers. However, little is known about the causes of the variation and how to reduce them.

Proprietary & Confidential. Health Care Incentives Improvement Institute, Inc. 5

Example: COPD Episodes Among Two Provider Groups

5

•  On average, Provider 10’s episode costs are 2.5x Provider 4’s.

•  How much of this difference is due to patient severity, provider competence and practice patterns?

Provider 4 Provider 10

N 408 444

Total Episode Costs $569,160 $1,640,580

Average Episode Costs $1,395 $3,695


Let’s split the average episode costs between typical and PAC

6

•  Provider 10 incurs more PACs on average relative to total annual episode cost (35% v. 25%)

•  On average Provider 10 incurs close to 4x more than Provider 4 on PACs per patient

Provider 4 Provider 10 Difference

Average PAC Costs $344 $1,297 3.8

Average Typical Costs $1,051 $2,398 2.3

% of Dollars Spent on PACs 25% 35% 1.4

3.8x

2.3x


PAC Cost and Frequency

7

Provider 10’s PAC problem is not caused by the frequency of PACs as much as the average cost per PAC. Let’s understand the drivers.


What types of PACs are driving excess costs for Provider 10?

8

PAC Type Provider 4 Provider 10 Type 1 PACs (directly related to the index condition) 41% 52% Type 2 PACs (related to comorbidities) 55% 36% Type 3 PACs (suggesting patient safety failures) 4% 12%

PAC Description Provider 4 Provider 10

Emergency Room 18% 12%

Acute Exacerbation of COPD 16% 15%

Pneumonia, Lung Complications 6% 13%

Cardiac Dysrhythmias 11% 9%

Urinary Tract Infections 4% 1%

Cellulitis, Skin Infections 8% 6%

Gastritis, ulcer, GI hemorrhage 7% 3%

Chronic Obstructive Bronchitis -- 6%

Complications of medical care, surgery, implanted device, grafts

2% 2%

DVT, Pulmonary Embolism 1% 8%


Is the variation in typical costs due to patient severity?

9


After adjusting for severity

•  After adjusting for patient severity, the difference in typical costs is minor and mostly due to practice pattern variation (e.g. higher use of discretionary services – labs, imaging, etc..)

Provider 4 Provider 10

Severity Index (reference = 1.0) 0.62 1.61

Average Typical after adjusting for severity $1,893 $2,227

1.2x


After adjustments, Provider 10 still has higher episode costs

11

However, the rank of the other providers has shifted – 3 went from low cost to relatively high cost, and 6 went from average to lower cost.


One view of performance: ranking on PACs and Typical

12

Provider 10 goes from being a complete outlier, to a Provider with a PAC problem, that can be addressed through clinical reengineering and improving patient management.


Observations

•  PAC Analysis and its details can help providers understand the source of professional variation and improve quality/cost performance

•  Severity-adjusting typical costs and comparing providers on that basis provides insights on practice pattern variations and price

•  Examining variation in episode cost can help to inform quality improvement initiatives, contracting decisions, and help providers and payers win in “budget-based” payments.




Contracting ECRs

IHA P4P Conference March 23rd, 2011

Doug Emery, M.S. Implementation Leader, Western Region HCI3


Agenda

•  ECR Concept and Accountable Care •  Retrospective vs Prospective Payment •  Negotiating Tunable Parameters


Patient-centered continuum of care

Source: Sustaining the Medical Home: How Prometheus Payment Can Revitalize Primary Care, RWJ publication


Amplitude of cost along the continuum

Source: Sustaining the Medical Home: How Prometheus Payment Can Revitalize Primary Care, RWJ publication


Retrospective Payment

5

Readmission

Hospitalization

Outpatient Professional

Key: Claims for typical care and services Potentially avoidable cost (PACs)

365 days – e.g. Jan 1.

Chronic Episode

Procedural Episode Inpatient

Professional

Outpatient Professional

180 day look-forward 30 day look-back


Retrospectively reconciled prospective episode budgets

Episode budget is established upon trigger and not paid out, but allocated

1

Claims incurred are paid FFS by the plan and accumulated, split between typical and PAC

2

Accumulated claims are compared to budgeted episode, and reconciled, to determine upside/ downside

3


Operational elements

•  Prospective budgets are calculated by the plan, and the tunable parameters (e.g. stop-loss, PAC allowance) are negotiated with providers. –  Need a contracting organization (health plan) –  Providers don’t have to be legally/organizationally integrated

•  Claims incurred are paid according to underlying FFS schedules and accumulated against budgets –  Need an accumulating engine (currently MedAssets, but

working with McKesson and TriZetto for additional solutions) •  Paid claims are reconciled against budgets and upside/

downside is determined –  Need fiscal intermediary (plan) to deduct withholds and

process reconciliations, pay out balances


Prospectively paid episode budgets with reconciliation

Episode budget is established upon trigger and paid out to “ACO” or fiscal intermediary

1

Claims incurred are zeroed out, but accumulated against budget

2

Claims incurred compared against budget to provide guidance on prospective budgets/payments

3


Operational elements

•  Prospective budgets are paid out (with or w/o withhold – we recommend always taking a withhold) –  Need ACO or fiscal intermediary who will distribute funds to

subcontracted providers •  Claims incurred are zeroed out but accumulated against

budgets to help inform future pricing and severity adjustment reconciliations –  Need claims engine to zero out claims for triggered budgets

and accumulate claims against episodes •  Incurred is reconciled against budgets, and withhold is then

balanced out (to guard against overpayment when patients die or seek services outside contracted providers) –  Need fiscal intermediary to perform reconciliations and

determine distributions of withhold balances


Negotiating Tunable Parameters


Risk Bifurcation

Total Relevant Costs of a Specific Episode

Total Cost of Care

Typical Costs of Episode Costs of all Potentially

Avoidable Complications (and other provider-specific variation)

Costs of all Base Services

Costs of all Severity Adjusters

Insurer – Probability risk Provider – Technical risk Consumer – Probability risk

Reliable Care

Global Cap

“Coarse” Episodes

ECRs

11


Technical Risk vs. Probability Risk

0 > EOC > 1 .0000017835

Insurance Market: Probability Risk where medical demand is uncertain

Typical Costs

PAC Costs

Catastrophic Costs EOC = 1

Medical Certainty

Tn Episode Conclude

T0 Episode Trigger

Risk Corridor = Negotiated Ratio between Typical vs PAC

Stop Loss


Negotiating Risk Corridors

Informed by guidelines and empirical data analysis

Adjusts ECR for local patterns

Arrived at through step-wise multi-variable regression model

Currently based at 10% of typical

Based on 50% of current defect rate

Total ECR price = Type of services * Frequency * Price per service

$3,600 -- $22,600

Typical Care

$360 -- $2,600

$3,000 -- $16,500

CHF ECR Range

$6,960 -- $41,700


Tunable Parameters: Negotiating ECR Risk Corridors

The PAC Allowance:

0% 100% 50%

CHF Prospective Budget @ 50% Patient 1 = $4,112 Patient 2 = $15,549 Patient 3 = $34,629

CHF Prospective Budget @ 75% Patient 1 = $5,348 Patient 2 = $18,714 Patient 3 = $41,013

14


Tunable Parameters: The Donut Hole

15

% o

f Cur

rent

PA

C C

osts

Re-

allo

cate

d to

E

CR

Bud

get (

PAC

Allo

wan

ce)

100%

50%

Size of “Donut Hole”

Payer Risk

Payer Risk

Provider Risk – “Donut Hole”


Tunable Parameters: The Donut Hole

16

% o

f Cur

rent

PA

C C

osts

Re-

allo

cate

d to

E

CR

Bud

get (

PAC

Allo

wan

ce)

100%

50%

Size of “Donut Hole”

Payer Risk

Payer Risk

Provider Risk – “Donut Hole”

Stop Loss


Prospective vs Retrospective Payment

Transparency only Programs

Fixed fee/bonus-based Programs

Shared savings-based Programs

Episode of Care payment

Provider Risk

Sav

ings



Documents

PROMETHEUS Pilot Pre- conference · Proprietary & Confidential. Health Care Incentives Improvement Institute, Inc. PROMETHEUS Pilot Pre-conference March 23rd 2011