Post Acute Care Services Are the Fastest Growing Category of - - PDF document

1

Post‐Acute Care: The Next Frontier for Controlling Medicare Spending

Robert Mechanic, MBA Brandeis University Estes Park Institute November 4, 2014

Post‐Acute Care Services Are the Fastest Growing Category

• f Medicare Spending

Brandeis University

2

Why Should Hospitals and Physicians Care?

2

Bundled Payment

Hospital or Integrated Network

$$$ Single payment to cover costs of episode of care (30, 60, 90 days)

Payer

$ $ $ $ $

Group is responsible for all care within the episode

Shared Accountability

90 day look‐forward Index Hospitalization Inpatient Professional Outpatient Professional

Professional services Inpatient Stays

Brandeis University

Readmission SNF

What’s in an Episode?

3

90 day look‐forward Index Hospitalization Inpatient Professional Outpatient Professional

Professional services Inpatient Stays

Brandeis University

Readmission SNF

CMMI Bundled Payment Pilot

Model 1

30 ‐ 90 day look‐forward Index Hospitalization Inpatient Professional Outpatient Professional

Professional services Inpatient Stays

Brandeis University

Readmission SNF

CMMI Bundled Payment Pilot

Model 2

4

30 day look‐forward Index Hospitalization Inpatient Professional Outpatient Professional

Professional services Inpatient Stays

Brandeis University

Readmission SNF

CMMI Bundled Payment Pilot

Model 3

30 day look‐forward Index Hospitalization Inpatient Professional Outpatient Professional

Professional services Inpatient Stays

Brandeis University

Readmission SNF

CMMI Bundled Payment Pilot

Model 4: Prospective Payment

5

Medicare Spends a Tremendous Amount in the 30 – 90 Days After Patients Are Discharged from the Hospital

9

Medicare Post Acute Care Spending

Hospital IP Professional Post‐Acute Hospital OP

2012 Medicare Spending by Type

21%

Source: MedPAC, 2014 Data Book (Charts 1‐1, 8‐2).

Hospital IP Professional Post‐Acute

2008 Medicare Spending for Hospitalization plus 30 Days

34%

Source: RTI Inc, Post‐Acute Care Episodes: Expanded Analytic File, June 2011 p.216.

Brandeis University

10

6

Medicare Payment Methods

• SNF: Per‐diem payment with therapies

billed separately

– Patients covered for up to 100 days

• Home health: 60‐day bundle
• Inpatient Rehab: Prospective per case

payment (similar to DRG method)

– 60 percent of patients must have one of 13 conditions

Brandeis University

11

• Avg. 2008 Medicare Inpatient Payments

for Select DRGs

$11,079 $5,347 $5,322 $6,437 $6,075 5,000 10,000 15,000 20,000 470 ‐ Maj. Joint 194 ‐ Pne w/CC 292 ‐ Heart Fail w/CC 683 ‐ Renal Failure w/CC 190 ‐ COPD w/MCC

Index Admission

12

Source: RTI Inc, Post‐Acute Care Episodes: Expanded Analytic File, June 2011

7

2008 Medicare Acute and Post‐Acute Payments for Inpatient‐Initiated 30‐Day Episodes

5,000 10,000 15,000 20,000 470 ‐ Maj. Joint 194 ‐ Pne w/CC 292 ‐ Heart Fail w/CC 683 ‐ Renal Failure w/CC 190 ‐ COPD w/MCC

Index Admission Post Acute

$18,414 $9,732 $10,636 $10,470 $12,456

13 Source: RTI Inc, Post‐Acute Care Episodes: Expanded Analytic File, June 2011. Thirty day fixed episodes include the full amount of all claims incurred within 30 days of discharge even if they extend beyond the 30 days period.

2008 Medicare Acute and Post‐Acute Payments for Inpatient‐Initiated 90‐Day Episodes

5,000 10,000 15,000 20,000 470 ‐ Maj. Joint 194 ‐ Pne w/CC 292 ‐ Heart Fail w/CC 683 ‐ Renal Failure w/CC 190 ‐ COPD w/MCC

Index Admission 30 day Post Acute 90 day Post Acute

$19,745 $12,479 $14,692 $14,910 $16,589

14

Source: RTI Inc, Post‐Acute Care Episodes: Expanded Analytic File, June 2011. 30‐90 day amounts are estimated based on RTI, Analysis of Acute Care Episode Definitions Chart Book, November 2009.

8

There is Significant Variation in Post‐Acute Care Spending Across Hospitals …. …. And Many Opportunities to Reduce Post‐Acute Care Spending

15

Average 2009 Post‐Acute Care Spending per Episode for Total Joint Replacement (90 day)

$0 $2,000 $4,000 $6,000 $8,000 $10,000 $12,000 $14,000 $16,000 A B C D E F G H I J K L M N O P Q R

16

Source: Brandeis University analysis of Medicare Claims data. Figures adjusted for hospital wage index.

$6,000 $12,000 “St. Minimus” “St. Maximus”

9

A Tale of Two Hospitals: Joint Replacement Episode

17

Source: Brandeis University analysis of Medicare Claims data. Unadjusted data.

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% Readmission Rate

• Pct. SNF
• Pct. Home Health
• St. Maximus
• St. Minimus

18

A Tale of Two Hospitals: Joint Replacement Episode

Source: Brandeis University analysis of Medicare Claims data.

10

Opportunities for St. Maximus

• Expand home health and reduce use of SNF

services where appropriate

• Put a program in place to monitor patients

following discharge

– Medication reconciliation – Home assessment – Primary care visit within 7 days – Emergency plan for likely events

• Consider preferred relationships with

collaborative & high value facilities.

19

Post Acute Strategy Components

• 1. Right setting
• 2. Right partners
• 3. Right relationships

– Patient & Family – Primary Care Physician – Post‐Acute Providers

Brandeis University

20

11

2008 Medicare Post‐Acute Care Payments Per User by Site of Service: DRG 470 (Total Joint)

$11,079 $3,132 $8,562 $12,596 $23,017 $9,496 $0 $5,000 $10,000 $15,000 $20,000 $25,000 Admission Home Health SNF Rehab LTAC Readmission

21

Source: RTI Inc, Post‐Acute Care Episodes: Expanded Analytic File, June 2011

100% 60% 40% 7% 0.2% 9%

Percent with Service: Within 30 Days of Hospital Discharge

Variation in 2010 Medicare Average Length of Stay for Skilled Nursing Facilities

10 20 30 40 50 60 Quartile 1 Quartile 2 Quartile 3 Quartile 4

22

Source: Adapted form Office of HHS Inspector General December 2010.

29 29 34 34 61 24 24 5 Difference Between Top & Bottom Quartile 10 Days = $4,000+

12

$15,961 $20,717 $9,336 $9,299 $7,929 $12,835 $0 $5,000 $10,000 $15,000 $20,000 $25,000 SNF A SNF B SNF C SNF D SNF E SNF F

23

2013 Average SNF Spending Per Admission for Total Joint Replacement Patients

One Large Hospital’s Top 6 SNFs by Number of Admissions

Source: Brandeis University analysis of Medicare claims data. All SNFs have 10+ cases.

Variation in 2009 Risk Adjusted Readmission Rates from Skilled Nursing Facilities

14.4% 18.1% 22.0%

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 25th Percentile Median 75th Percentile

Readmissions

24

Source: MedPAC Report to Congress, March 2012.

13

Expert Panel Ratings of Whether Hospital Admissions from Nursing Home Were Avoidable NH Resident Group Yes No Medicare (n=94) 69% 31% Medicaid/Other (n=106) 65% 35% High Readmit NHs (n=101) 75% 25% Low Readmit NHs (n=99) 59% 41% All Residents (n=200) 67% 33%

25

Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 58:627‐635, 2010.

Causes of Potentially Avoidable Admissions

Diagnoses for Potentially Avoidable Admissions n=100

Cardiovascular (mostly CHF, chest pain) 22% Respiratory (mainly pneumonia, bronchitis) 21% Mental status change 13% Urinary tract infection 11% Sepsis or fever 8% Skin (cellulitis, wound, pressure ulcer) 8% Dehydration 7% Gastrointestinal (bleeding, diarrhea) 7%

Brandeis University

26

Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 58:627‐635, 2010.

14

Ratings of Factors Associated with Potentially Avoidable Admissions

Factors that could have prevented hospitalization

Important Somewhat Important

NH should have been able to do everything done by hospital

50% 34%

Better quality by NH physician or AP

55 28

One MD visit could have prevented transfer

37 43

Better quality by NH staff

24 48

Better advance care planning

38 24

Resident’s condition limited ability to benefit from the transfer

19 28

Brandeis University

27

Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 58:627‐635, 2010.

Resources Rated as Potentially Helpful in Preventing Avoidable Admissions

Would Prevent Transfer Very or Somewhat Helpful

Exam by MD or APC within 24 hrs 40% 52% MD or APC in NH 3 days/week 16 80 RN providing care vs. LPN or Aide 6 85 IV therapy available in NH 22 58 Lab tests within 3 hours 15 74

Brandeis University

Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 58:627‐635, 2010.

15

29

Assessing Quality is Difficult

Developing a Preferred SNF Network

Criteria Criteria

Right Geography Right Geography Willing to collaborate

• n QI

Willing to collaborate

• n QI

Strong Performance Metrics Strong Performance Metrics

On‐Site MD Coverage On‐Site MD Coverage Ability to Manage Complex Patients Ability to Manage Complex Patients

Brandeis University Source: Atrius Health.

16

Performance in One Health System’s Preferred SNF Network

Brandeis University

31

15 21 32

5 10 15 20 25 30 35 Medicare Advantage Pioneer ACO Market Avg.

Average Length of Stay 6 Days = $2,400+

Source: Atrius Health, 2013.

Performance Expectations

• Appropriate staffing ‐ low staff turnover
• Able to manage complex patients
• Able to treat acute exacerbations in place
• Use “preferred” MD/APC provider with

24/7 coverage

Brandeis University

32

17

Performance Expectations

• Committed to collaborative QI work
• Point person for clinical communication
• Discharge planning begins at admission
• Regular performance reporting
• Use preferred vendors (HHA, DME etc.)

Brandeis University

33

Emerging Innovations

34

18

Emerging Innovations

Brandeis University 35

Emerging Innovations

Brandeis University 36

19

Questions

Robert Mechanic The Heller School for Social Policy & Management The Health Industry Forum Brandeis University mechanic@brandeis.edu www.healthforum.brandeis.edu

37