6/30/10 Lieberman Consulting Lieberman Consulting Inc Inc Table - - PDF document

6 30 10
SMART_READER_LITE
LIVE PREVIEW

6/30/10 Lieberman Consulting Lieberman Consulting Inc Inc Table - - PDF document

6/30/10 Lieberman Consulting Lieberman Consulting Inc Inc Table of Contents IL-ACC Practice Management Symposium Accountable Care Organizations Unsustainable Non-System Rewards Quantity, Not Quality What Are ACOs? ACO Core Principles Permit


slide-1
SLIDE 1

6/30/10 1

1

Lieberman Consulting Inc

IL-ACC Practice Management Symposium Accountable Care Organizations

What Are ACOs? How Might They Work?

Steve Lieberman

May 14, 2010 Chicago, IL

2

Lieberman Consulting Inc

Table of Contents

ACO Core Principles Permit Diverse Arrangements Patient Attribution, Shared Savings, Quality Measurement Will ACOs Work? Challenges & Opportunities Unsustainable Non-System Rewards Quantity, Not Quality

3

Lieberman Consulting Inc

FFS Medicare: Unsustainably Expensive & Unfair

“Here … a medical community came to treat patients the way subprime mortgage lenders treated home buyers: as profit centers.” Atul Gawande 2006 Spending 92-06 Growth McAllen $14,946 8.3% La Crosse $5,812 3.9%

Gawande, Atul. The Cost Conundrum: What a Texas town can teach us about health care. The New Yorker, June 1, 2009

“the federal budget is on an unsustainable path . . . rising costs for health care . . . will cause federal spending to increase rapidly under any plausible scenario . . .” (The Long-Term Budget Outlook, CBO, 2009)

4

Lieberman Consulting Inc

Growth in Per-Capita Medicare Spending

Average annual inflation adjusted growth rates, 1992-2006

4 .5% to 8 .4% (59) 4 .0% to < 4 .5% (52) 3 .5% to < 4 .0% (68) 3 .0% to < 3 .5% (62) 1 .6% to < 3 .0% (65) Not Populated 5

Lieberman Consulting Inc

Source: Slowing the Growth of Health Care Spending: Lessons from Regional Variation Fisher, Skinner, Bynum, New England Journal of Medicine, February 26, 2009

A Tale of 5 Cities

6

Lieberman Consulting Inc

Three-fold Variation in Medicare Spending

Per Capita Medicare Spending, 2007 (Age-Sex-Race-Adjusted)

$10,250 to 17,184 (55) 9,500 to < 10,250 (69) 8,750 to < 9,500 (64) 8,000 to < 8,750 (53) 6,039 to < 8,000 (65) Not Populated

slide-2
SLIDE 2

6/30/10 2

7

Lieberman Consulting Inc

What do higher spending regions get?

8

Lieberman Consulting Inc

What do higher spending regions get?

(1) ¡Fisher ¡et ¡al. ¡Ann ¡Intern ¡Med: ¡2003; ¡138: ¡273-­‑298 ¡ ¡ (2) ¡Baicker ¡et ¡ ¡al. ¡Health ¡Affairs ¡web ¡exclusives, ¡October ¡ ¡7, ¡2004 ¡ (3) ¡Fisher ¡et ¡al. ¡Health ¡Affairs, ¡web ¡exclusives, ¡Nov ¡16, ¡2005 ¡ (4) ¡Skinner ¡et ¡al. ¡Health ¡Affairs ¡web ¡exclusives, ¡Feb ¡7, ¡2006 ¡ (5) ¡Sirovich ¡et ¡al ¡Ann ¡Intern ¡Med: ¡2006; ¡144: ¡641-­‑649 ¡ (6) ¡Fowler ¡et ¡al. ¡JAMA: ¡299: ¡2406-­‑2412 ¡

Health Outcomes

Slightly higher mortality

Patient-Perceived Quality

Worse access to primary care Lower overall rating

  • f medical care

Lower satisfaction with hospital care

Physician’s Perceptions

Worse communication among physicians Greater difficulty ensuring continuity Lower satisfaction with career

Trends Over Time

Greater growth in per capita resource use Lower gains in survival following AMI

9

Lieberman Consulting Inc

Variations by Types of Care: Evidence-Based Care

Includes ¡Rates ¡of: ¡ ¡ Mammogram, ¡Women ¡65-­‑69 ¡ ¡Pap ¡Smear, ¡Women ¡65+ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ Pneumococcal ¡ImmunizaYon ¡ ¡ ¡ ¡Aspirin ¡at ¡admission ¡(Heart ¡aZack) ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

Evidence-Based Quality

2.0 1.0 0.5 More Care in High Spending Regions Less Care in High Spending Regions

10

Lieberman Consulting Inc

Variations in Preference Sensitive Care

Includes ¡rates ¡of: ¡ Total ¡Hip ¡Replacement ¡Total ¡Knee ¡Replacement ¡ Back ¡Surgery ¡ ¡CABG ¡Following ¡Heart ¡AZack ¡

Evidence-Based Quality Preference Sensitive Care

2.0 1.0 0.5 More Care in High Sending Regions Less Care in High Spending Regions

11

Lieberman Consulting Inc

Preference Sensitive Care Varies in ALL Regions

2.0 4.0 6.0 8.0 10.0 12.0 14.0 Q1 Q2 Q3 Q4 Q5 Rate of Coronary Artery Bypass Graft Surgery

Age-sex-race adjusted, 2001

Rate per 1000 Enrollees HRRs by Spending Quintile

Each red dot represents a Hospital Referral Region (HRR) Each red dot represents a Hospital Referral Region (HRR)

12

Lieberman Consulting Inc

Variation in Supply Sensitive Care

Includes ¡rates ¡of: ¡ InpaYent ¡Days ¡ ¡InpaYent ¡Days ¡in ¡the ¡ICU ¡ Imaging ¡& ¡DiagnosYc ¡Tests ¡EvaluaYon ¡& ¡Management ¡Visits ¡

Evidence-Based Quality Preference Sensitive Care

2.0 1.0 0.5

Supply Sensitive Care

More Care in High Spending Regions Less Care in High Spending Regions

N.B. ¡Self-­‑reported ¡health ¡ status ¡& ¡income ¡explain ¡ about ¡25% ¡of ¡variaYon ¡

slide-3
SLIDE 3

6/30/10 3

13

Lieberman Consulting Inc

Table of Contents

Unsustainable Non-System Rewards Quantity, Not Quality Patient Attribution, Shared Savings, Quality Measurement Will ACOs Work? Challenges & Opportunities ACO Core Principles Permit Diverse Arrangements

14

Lieberman Consulting Inc

Comparison of Different Payment Models

FFS Capitation ACO

Payment Providers are revenue centers rewarded for increased volume With fixed payments unrelated to volume, providers are cost centers rewarded for “stinting” Moderating incentives, shared savings balances incentives of FFS & capitation Patients Neither assigned nor enrolled Enrolled with specific provider Assigned based on previous care patterns Primary care & care coordination Little reward for primary care or care coordination Supports primary care and care coordination Supports primary care and care coordination Accountability for per-capita costs & quality Weak incentives to manage per-capita costs or improve quality Strong accountability for per- capita cost but no link to quality Accountability for costs. Links shared savings to meeting quality measures 15 Lieberman Consulting Inc

Accountability, “Systemness” & Incentives

  • Establish robust HIT infrastructure
  • Implement cost-saving and quality-

improving medical interventions

  • Evaluate performance of systems
  • Restructure payment incentives to

avoid extremes of FFS “revenue centers” & capitation “cost centers” Key Design Elements

  • Pay for better value: improve
  • verall health & reduce costs
  • Tools: timely feedback to providers
  • Reporting: require utilization and

quality data from providers New model: It’s the system - Establish

  • rganizations accountable for aims and

capable of redesigning practice and managing capacity Realign incentives – both financial and clinical – to support accountability for costs and quality across care settings Core Principles Achieve better health, better quality & lower costs for patients and communities Better information that engages physicians, supports improvement, and informs consumers

16

Lieberman Consulting Inc

ACOs Differ But Share a Few, Key Elements

Can provide or manage continuum of care as a real or virtually integrated delivery system Are of a sufficient size to support comprehensive performance measurement Are capable of internally distributing shared savings payments

1 2 3

Important Caveats

  • ACOs are not gatekeepers
  • ACOs do not require changes to benefit structures
  • ACOs do not require patient enrollment

17

Lieberman Consulting Inc

Performance Payment Framework

  • ACO receives mix of FFS and

prospective fixed payment

  • If successful at meeting

budget and performance targets, greater financial benefits

  • If ACO exceeds budget, more

risk means greater financial downside

  • Only appropriate for providers

with robust infrastructure, demonstrated track record in finances and quality and providing relatively full range

  • f services
  • Payments can still be tied

to current payment system, although ACO could receive revenue from payers and distribute funds to members (depending on ACO contracts)

  • At risk for losses if spending

exceeds targets

  • Increased incentive for

providers to decrease costs due to risk of losses

  • Attractive to providers with

some infrastructure or care coordination capability and demonstrated track record

  • Continue operating under

current insurance contracts/ coverage models (e.g., FFS)

  • No risk for losses if spending

exceeds targets

  • Most incremental approach

with least barriers for entry

  • Attractive to new entities,

risk-adverse providers, or entities with limited

  • rganizational capacity, range
  • f covered services, or

experience working with

  • ther providers

Level 1 Asymmetric shared-savings Level 2 Symmetric Model Level 3 Partial Capitation Model

ACOs offer a wide range of approaches

18

Lieberman Consulting Inc

Advanced

  • ACOs use more complete

clinical data (e.g., electronic records, registries) and robust patient-generated data (e.g., Health Risk Appraisals, functional status)

  • Well-established and

robust HIT infrastructure

  • Focus on full spectrum of

care and health system priorities Intermediate

  • ACOs use specific clinical

data (e.g., electronic laboratory results) and limited survey data

  • More sophisticated HIT

infrastructure in place

  • Greater focus on full

spectrum of care Beginning

  • ACOs have access to

medical, pharmacy, and laboratory claims from payers (claims-based measures)

  • Relatively limited health

infrastructure

  • Limited to focusing on

primary care services (starter set of measures)

Multiple priorities, outcome-oriented, and span the continuum of care

Quality Measures Will Evolve Over Time:

Beginning, Intermediate, & Advanced Stages

slide-4
SLIDE 4

6/30/10 4

19

Lieberman Consulting Inc

Payer/Provider Requirements for an ACO

  • Attribution algorithm assigns patients to providers (if unique to ACO)
  • PMPM budgets for assigned patients reflect past spending (& trend)
  • Payer shares savings with ACO, if financial & quality targets met
  • Meaningful measurement needs minimum “risk pool” size (e.g.,

15,000 commercial, 10,000 Medicaid, or 5,000 Medicare patients)

  • Accept common core set of quality measures (minimum; can add)
  • Payers provide ACO timely data
  • Payers & providers cooperate to lower costs & improve quality

20

Lieberman Consulting Inc

Table of Contents

Unsustainable Non-System Rewards Quantity, Not Quality ACO Core Principles Permit Diverse Arrangements Will ACOs Work? Challenges & Opportunities Patient Attribution, Shared Savings, Quality Measurement

21

Lieberman Consulting Inc

Goals of Patient Assignment Method

Important Caveats

  • Patients assigned by plurality of outpatient E&M visits (PCP-1st; Medical Specialist-2nd;

Surgical Specialist-3rd)

  • For patient assignment, PCPs must be exclusive to one ACO (to minimize concerns

about selection & dumping); Specialists can be part of multiple ACOs

  • The method is not meant to establish individual provider accountability
  • Accountability for assigned patients lies with the ACO, not the individual provider
  • Physicians are part of the ACO system of care
  • Even providers affiliated with only one ACO can refer patients to non-ACO providers

Unique provider assignment for every patient (no enrollment by patients) No “lock in” of patients to the ACO (not a gatekeeper model) Patients are assigned based on where they received their care in the past Minimize “dumping” of high risk or high cost patients

22

Lieberman Consulting Inc

Understanding Provider Relationship

Community Providers Accountable Care Organizations Bonus- Eligible Providers

(ACO defined)

Community Providers not part of ACO but may provide care to ACO patient. Some community providers may contract with ACO or routinely receive referrals, while others may have no relationship (or be out of area). ACO Providers: Members govern ACO and, if exclusive, have patients assigned to them. Other providers may join multiple ACOs. Bonus-Eligible Providers: ACO prospectively sets eligibility and allocates shared savings. ACOs have discretion to pay bonuses to a subset or all ACO members, varying treatment and amounts (e.g., all PCPs could receive bonuses, while only some specialists might).

Providers Used for Patient Assignment

(ACO Defined)

23

Lieberman Consulting Inc

Savings Based on Spending Targets

Projected Spending Actual Spending

Shared Savings

Target Spending

ACO Launched

24

Lieberman Consulting Inc

Current ACO Model Impact

Level of Measurement Individual ACO (System-Level) Reduces fragmentation and silos of practice; and, provides an assessment of care because many providers contribute to a patient’s care over time. Types of Measures Process Outcomes, Patient Experience, Efficiency Better data for patients to make choices about providers better data for providers to make changes; Increased accountability for resource use. Measurement Focus Individual Provider Accountability for Process Care Coordination, Shared Decision Making, Capacity Control Organizational support for managing and improving care; better patient engagement Provider Focus Discrete Patient Encounters Overall health of the population Shared accountability for the continuum of care.

Measuring Performance in ACOs

slide-5
SLIDE 5

6/30/10 5

25

Lieberman Consulting Inc

Starter Set of ACO Quality Measures

Domain Beginning Measures*

Overuse Low back pain: use of imaging studies Appropriate Testing for Children With Pharyngitis Avoidance of Antibiotic Treatment in Adults with Acute Bronchitis Appropriate treatment for children with upper respiratory infection (URI) Population Health Breast Cancer Screening Cervical Cancer Screening Colorectal Cancer Screening Diabetes: HbA1c Management (Testing) Diabetes: Cholesterol Management (Testing) Cholesterol Management for Patients with Cardiovascular Conditions (Testing) Use of appropriate medications for people with asthma Persistence of Beta-Blocker Treatment After a Heart Attack Safety Annual monitoring for patients on persistent medications Care Coordination** 30-day all cause (risk-adjusted) readmission rate * Most are HEDIS measures ** Test measure 26 Lieberman Consulting Inc

Table of Contents

Unsustainable Non-System Rewards Quantity, Not Quality ACO Core Principles Permit Diverse Arrangements Will ACOs Work? Challenges & Opportunities Patient Attribution, Shared Savings, Quality Measurement

27

Lieberman Consulting Inc

ACO Pilot Sites

Carilion Clinic Roanoke, VA

  • ~900 Providers
  • 60,000 Medicare

Patients Assigned Norton Healthcare Louisville, KY

  • ~400 Providers
  • 30,000 Medicare

Patients Assigned Tucson Medical Center Tucson, AZ

  • ~80 Providers
  • 10,000 Medicare

Patients Assigned

Low Competitive Highly Competitive Environment Environment Fully Integrated Multiple Independent System Provider Groups Large Group Small Group

28

Lieberman Consulting Inc

Key Challenges for ACOs

  • Will “critical mass” of providers join, with enough assigned patients?
  • Will payers support Level I ACOs, or only focus on integrated systems

ready for Level II or III?

  • Financing for ACO start-up costs? (e.g., Infrastructure, IT, analysis,

limiting ER use, etc.)

  • Are performance measures, patient assignment algorithm, and

budgeting methodology “good enough” to get started? How to improve?

  • Can ACOs change patient behavior & provider culture without

enrollment, “lock-in”, change in benefits & modest (Level I) incentives?

  • Potential to increase provider concentration and power?

29

Lieberman Consulting Inc

Why ACOs Might Succeed (Over Time)

  • Broad, flexible system built on essential core principles

– Lots of local variation possible within ACO concept

  • 3 ACO Levels permit tailoring to different circumstances

– “Training Wheels” for Level I entities (no risk) – Level II offers more reward but adds (limited) risk – Partial Capitation allows proven entities to add FFS Medicare & PPOs

  • Pathway to begin fundamental shift from FFS to population health

& accountable care

  • Opportunity for providers to change clinical & business environment

– Requires timely data, analysis & working as a system of care

30

Lieberman Consulting Inc

Supplemental Background

  • Regional Variations
  • Patient Assignment Algorithm
  • ACO Financial Targets
  • Quality Measures
slide-6
SLIDE 6

6/30/10 6

31

Lieberman Consulting Inc

Standardized Per-Capita Medicare Payments 2007

$9,300 to 15,750 (55) 8,600 to < 9,300 (65) 7,900 to < 8,600 (56) 7,200 to < 7,900 (62) 5,279 to < 7,200 (68) Not Populated 32

Lieberman Consulting Inc

Effects of Adjusting & Not Adjusting for Price

33

Lieberman Consulting Inc

Patient Assignment Methodology

2-Years of Claims for Eligible Patients Outpatient, E&M Services w/ allowed charges > $0 List of Provider Eligible for Assignment Assigned a Specialty “Bucket”

  • 1. Primary

Care

  • 2. Medical

Specialist

  • 3. Surgical

Specialist Patient Assigned to Provider* with Highest Number of E&M Visits

  • If tie, provider with longest time

between first and last visit is assigned

  • If a tie at 1 visit, provider with most

recent visit gets assigned * PCP(1st), Medical Specialist(2nd), Surgical Specialist(3rd)

34

Lieberman Consulting Inc

Key Considerations for Spending Benchmarks

1. Setting expenditure target for ACO 2. Savings Threshold Necessary in Asymmetric Risk (no downside) model – Limits bonus payments for savings that happen by chance 3. Percentage of Savings to be Shared by Providers and Payers – Negotiated and influenced by type of risk, threshold, etc. 4. Symmetric or Asymmetric Risk – Asymmetric (One-Sided) Model

  • ACOs not at risk for any increased costs if actual spending exceeds

benchmark amounts

  • ACOs share in savings if actual spending is below the benchmark spending

– Symmetric (Two-Sided) Model

  • ACOs share in the costs if exceed benchmark spending amounts
  • ACOs may be eligible for greater, first-dollar savings than one-sided models

5. Start-up funding (to finance ACO infrastructure, etc.)

35

Lieberman Consulting Inc

Bonus Threshold Model (Asymmetric Model)

  • Project benchmark ($469) in

the performance period based on historical baseline data

  • ACO reduced PMPM spending

by 5% ($24) from benchmark.

  • Exceeded 2% ($9) savings

threshold resulting in $15 in shared savings

  • If 50/50 shared savings,

$7.50 (1.5%) would be distributed to ACO, with $7.50 to payer

  • Note: No down-side risk if

actual spending exceeds benchmark

36

Lieberman Consulting Inc

Bonus Withhold Model (Symmetric Model)

  • ACOs could opt to have a

threshold or not (benefit is access to first dollar of savings; tradeoff is risk of first dollar losses and statistical volatility)

  • ACO shares downside risk
  • May want to use “risk

corridors”

  • Assume ACO exceeded the

benchmark by 10% ($46)

  • Without risk corridors, ACO

responsible for 80% of surplus/deficit ($37)

  • With risk corridor, losses/

gains limited ($18, if limited to 80% of +/- 5%)

slide-7
SLIDE 7

6/30/10 7

37

Lieberman Consulting Inc

Advancing Measurement in the ACO

  • Additional Testing/Surveillance Measures for Year 1

– Clinical-Enhanced, Claims-Based – Inpatient, Claims-Based – Utilization Measures

  • ER admissions
  • Admissions for Ambulatory-Sensitive Conditions
  • Use of generics
  • Testing in Year 2

– Patient Experience (CAHPS) – Care Transitions – Patient-Reported Health Status