The Future of Post-Acute Care Under Value-Based Payment Robert Mechanic, MBA Brandeis University Northeast Home Health Leadership Summit January 22, 2015
Medicare Margins for Freestanding Home Health Agencies 20.0% 15.0% 10.0% 5.0% 18.2% 19.4% 15.0% 14.4% 14.8% 12.6% 0.0% 2009 2010 2011 2012 2013 Proj. 2014 Proj. Source: MedPAC Report to Congress, March 2014. 2
Growth in Medicare Home Health Episodes and Payments 2002-12 2002 2012 Change Medicare enrollees 35.0 37.1 6% HHA Users (millions) 2.5 3.4 37% Episodes (millions) 4.1 6.7 64% Payments (billions) $9.6 $18.0 89% vs. Source: MedPAC Report to Congress, March 2014. 3 47% All Medicare Spending
Post-Acute Care Services Are the Fastest Growing Category of Medicare Spending Why Should Hospitals and Physicians Care? 4 Brandeis University
Thinking About Post-Acute Care 5 Brandeis University
20 18 16 14 12 10 8 6 4 2 0 Change is Coming Medicare Beneficiaries in Risk-Based Programs in 2014 2015 BPCI Applications 963 MD Groups 1,053 Hospitals 3,300 Post Acute Providers??? 5.7 15.7 Bundled Payment Medicare ACOs Medicare Advantage 6
Lets Talk About Disruptive Innovation 7
8 Million People Signed up for Exchange Coverage in 2014 $164 Billion CBO: Cost of ACA Premium Subsidy Will Fall by 14% Due to Lower Health Insurance Costs Disruptive enough? 8 Brandeis University
Change in Coverage Under the ACA 2014 2015 2016 Insurance Exchanges 6 13 24 Medicaid & CHIP 7 11 12 Employer & Non-group (1) (5) (11) Uninsured (12) (19) (25) Source: Congressional Budget Office, April 2014 Baseline 9 Brandeis University
Narrow Network Health Plans Dominate Many Individual Exchanges 10
Narrow Networks Have Substantially Lower Premiums 11
Now Republicans Control The Congress Does Anyone Remember the Ryan Plan? 12 Brandeis University
But Lets Get Back to Value- Based Payment 13 Brandeis University
Bundled Payment Payer $$$ Single payment to cover costs of episode of care (30, 60, 90 days) Shared Accountability Hospital or Integrated Network $ $ $ $ $ Group is responsible for all care within the episode
What s in an Episode? Outpatient Professional Professional services Inpatient Professional Inpatient Stays Index Hospitalization SNF Readmission 90 day look-forward Brandeis University
CMMI Bundled Payment Pilot Model 1 Outpatient Professional Professional services Inpatient Professional Inpatient Stays Index Hospitalization SNF Readmission 90 day look-forward Brandeis University
CMMI Bundled Payment Pilot Model 2 Outpatient Professional Professional services Inpatient Professional Inpatient Stays Index Hospitalization SNF Readmission 30-90 day look-forward Brandeis University
CMMI Bundled Payment Pilot Model 3 Outpatient Professional Professional services Inpatient Professional Inpatient Stays Index Hospitalization SNF Readmission 30 day look-forward Brandeis University
CMMI Bundled Payment Pilot Model 4: Prospective Payment Outpatient Professional Professional services Inpatient Professional Inpatient Stays Index Hospitalization SNF Readmission 30 day look-forward Brandeis University
Medicare Spends a Tremendous Amount in the 30 90 Days After Patients Are Discharged from the Hospital 20
Medicare Post Acute Care Spending 2012 Medicare Spending by Type 21% 2008 Medicare Spending for Hospitalization plus 30 Days 34% Hospital IP Post-Acute Professional Hospital OP Hospital IP Professional Post-Acute Source: MedPAC, 2014 Data Book (Charts 1-1, 8-2). 21 Source: RTI Inc, Post-Acute Care Episodes: Expanded Analytic File, June 2011 p.216. Brandeis University
Medicare 2012 Post-Acute Spending by Setting HHA = $18.3b 31% Source: MedPAC, 2014 Data Book. Home Health SNF IRF LTAC 22 Brandeis University
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 23 Brandeis University
Avg. 2008 Medicare Inpatient Payments for Select DRGs 20,000 15,000 10,000 5,000 0 $11,079 $5,347 $5,322 $6,437 $6,075 470 - Maj. Joint 194 - Pne w/cc 292 - Heart Fail w/cc Index Admission 683 - Renal Failure w/cc 190 - COPD w/mcc Source: RTI Inc, Post-Acute Care Episodes: Expanded 24 Analytic File, June 2011
2008 Medicare Acute and Post-Acute Payments for Inpatient-Initiated 30-Day Episodes 20,000 $18,414 15,000 10,000 $9,732 $12,456 $10,636 $10,470 5,000 0 470 - Maj. Joint 194 - Pne w/cc 292 - Heart Fail w/cc Index Admission Post Acute 683 - Renal Failure w/cc 190 - COPD w/mcc 25 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 20,000 15,000 $19,745 $12,479 $16,589 $14,692 $14,910 10,000 5,000 0 470 - Maj. Joint 194 - Pne w/cc 292 - Heart Fail w/cc 683 - Renal Failure w/cc Index Admission 30 day Post Acute 90 day Post Acute 190 - COPD w/mcc 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. 26
There is Significant Variation in Post-Acute Care Spending Across Hospitals.. And Many Opportunities to Reduce Post-Acute Care Spending 27
Average 2009 Post-Acute Care Spending per Episode for Total Joint Replacement (90 day) $16,000 $14,000 $12,000 $12,000 $10,000 $8,000 $6,000 $6,000 $4,000 $2,000 $0 A B C D E F G H I J K L M N O P Q R St. Minimus Source: Brandeis University analysis of Medicare 28 Claims data. Figures adjusted for hospital wage index. St. Maximus
A Tale of Two Hospitals: Joint Replacement Episode 29 Source: Brandeis University analysis of Medicare Claims data. Unadjusted data.
A Tale of Two Hospitals: Joint Replacement Episode 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% St. Maximus St. Minimus 20.0% 10.0% 0.0% Readmission Rate Pct. SNF Pct. Home Health Source: Brandeis University analysis of Medicare Claims data. 30
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. 31
Post Acute Strategy Components 1. Right setting 2. Right partners 3. Right relationships Patient & Family Primary Care Physician Post-Acute Providers 32 Brandeis University
2008 Medicare Post-Acute Care Payments Per User by Site of Service: DRG 470 (Total Joint) $25,000 Within 30 Days of Hospital Discharge $20,000 $15,000 $10,000 $23,017 $5,000 $0 Percent with Service: $11,079 $3,132 $8,562 $12,596 $9,496 Admission Home Health SNF Rehab LTAC Readmission 100% 60% 40% 7% 0.2% 9% 33 Source: RTI Inc, Post-Acute Care Episodes: Expanded Analytic File, June 2011
Variation in 2010 Medicare Average Length of Stay for Skilled Nursing Facilities 60 50 40 30 20 10 Difference Between Top & Bottom Quartile 10 Days = $4,000+ 24 5 29 24 34 29 61 34 0 Quartile 1 Quartile 2 Quartile 3 Quartile 4 Source: Adapted form Office of HHS Inspector General 34 December 2010.
2012 Geographic Variation in Per Capita Medicare Payments to Home Health Agencies Data are Price, Race, Age & Sex Adjusted $2,000 $1,500 $1,000 $2,133 $1,891 $1,324 US Average $500 $1,122 $0 $906 $627 $553 $481 $410 $370 $250 Source: Dartmouth Atlas of Health Care 35
2013 Average SNF Spending Per Admission for Total Joint Replacement Patients $25,000 One Large Hospital s Top 6 SNFs by Number of Admissions $20,000 $15,000 $10,000 $5,000 $15,961 $20,717 $9,336 $9,299 $7,929 $12,835 $0 SNF A SNF B SNF C SNF D SNF E SNF F Source: Brandeis University analysis of Medicare claims data. All SNFs have 10+ cases. 36
Variation in 2009 Risk Adjusted Readmission Rates from Skilled Nursing Facilities 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 14.4% 18.1% 22.0% 0.0% 25th Percentile Median 75th Percentile Readmissions Source: MedPAC Report to Congress, March 2012. 37
40.0% 35.0% 30.0% 25.0% Distribution of Readmission Rates Across Home Health Agencies in 2010 20.0% 15.0% 10.0% 25% 29% 39% 5.0% 0.0% 25th Percentile Median 75th Percentile Source: MedPAC Report to Congress, March 2012. 38
The Congress should direct the Secretary to reduce payments to home health agencies with high risk-adjusted rates of hospital readmission MedPAC 2014 39
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% Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 58:627-635, 2010. 40
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% Source: Ouslander et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes and costs. JAGS 41 58:627-635, 2010. Brandeis University
Ratings of Factors Associated with Potentially Avoidable Admissions Factors that could have prevented hospitalization NH should have been able to do everything done by hospital Important Somewhat Important 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 Source: Ouslander et al. Potentially avoidable hospitalizations of nursing 42 home residents: frequency, causes and costs. JAGS 58:627-635, 2010. 19 28 Brandeis University
Assessing Quality is Difficult 43
Developing a Preferred Post-Acute Network Right Geography Ability to Manage Complex Patients Criteria Willing to collaborate on QI Source: Atrius Health. On-Site MD Coverage Strong Performance Metrics Brandeis University
Performance in One Health System s 35 30 25 Preferred SNF Network Average Length of Stay 21 32 6 Days = $2,400+ 11 Days =$4,400+ 20 15 10 5 0 15 Medicare Advantage Pioneer ACO Market Avg. Source: Atrius Health, 2013. 45 Brandeis University
Performance Expectations Appropriate staffing - low staff turnover Able to manage complex patients Able to treat acute exacerbations in place Close linkage with preferred MD/APC Committed to collaborative QI work Point person for clinical communication Regular performance reporting 46 Brandeis University
Innovations 47 Brandeis University
Innovations 48
Innovations 49 Brandeis University
Innovations 50
Observations about Home Health Home health will not be a viable fee-for-service business for much longer Unit cost superseded by total medical expense Home health can be an important lever for success in value-based payment Agencies will need size and adequate capital investment to play in the new world Partnerships are critical; but many will struggle with how to best use you 51 Brandeis University
Questions Robert Mechanic The Heller School for Social Policy & Management The Health Industry Forum Brandeis University mechanic@brandeis.edu www.healthforum.brandeis.edu 52