Transcript Traci La Valle - HFMA Maryland

Maryland Value-Based Performance Policies
What we’ve achieved and where we’re headed
Traci La Valle
Vice President
Maryland Hospital Association
January 27, 2017
Today’s Agenda
• Big picture: Maryland performance
• Value-based performance policies
• Population health metrics
• Population-based measurement
2
The Maryland Demonstration is Designed
to Accomplish the Triple Aim
Maryland entered into a five-year contract with CMS to implement the
Maryland All-Payer Model, a demonstration under Section 1115A of
the Social Security Act, intended to test how global budgets could be
used to achieve The Triple Aim
The Triple Aim:
• Improve the experience of care for patients
• Improve the health of whole populations
• Reduce the per capita cost of health care
Better Health
3
How It Works
• Providing fixed, predictable revenues gives hospitals flexibility to invest in care and
health improvement activities such as:
— Coordinating care with other health care providers
— Reducing complications
— Reducing readmissions
• In turn, these activities reduce avoidable utilization, which improves value and
affordability
Physicians
Medications
Outpatient
clinics
Home health
Under the demonstration, Maryland’s
hospitals are accountable for driving
down health care costs, no matter
where or how the care is delivered
Nursing homes
Durable medical
equipment
4
Financial and Quality Performance
As a result of the demonstration, Maryland’s hospitals have saved Medicare over
$400 million, more than 3 times the amount that was required by this point in time
5
Reducing Medicare Costs
For the rest of the country, Medicare costs per beneficiary are increasing,
while in Maryland these costs are decreasing
Average Medicare Costs Per-Beneficiary-Per-Year
2013
2016
Change
Maryland
$10,548
$10,471
-0.7%
Nation
$9,280
$9,305
0.3%
Source: Maryland figures are calculated from 100 percent Medicare Parts A and B claims data. National
figures calculated from 5 percent sample file of Medicare claims
6
Bending the Cost Curve for Medicare
Medicare Spending per Beneficiary by Provider Type
June 2015 to June 2016 Percent Change
Hospital Spending
Inpatient
Outpatient
Total Hospital Spending
Non-Hospital Spending
SNF
Home health
Hospice
Outpatient Clinic & Other
Physician Professional
Claims
Total Non-Hospital
Spending
Total
Maryland
Nation
-2.57%
3.36%
-0.69%
0.50%
5.35%
1.96%
Dollar Impact of
Savings (Millions)
(52.0)
(13.9)
($65.9)
-6.12%
2.80%
8.73%
0.08%
-5.10%
-0.64%
2.16%
1.11%
(3.4)
4.9
5.7
(0.6)
4.11%
1.83%
24.5
2.17%
0.40%
$31.1
0.60%
1.16%
($34.8)
The demonstration drives a decrease in Medicare spending, in part
by incentivizing a shift in utilization to lower-cost settings of care
7
Monthly Case-Mix Adjusted Readmission Rates
16%
2014
2013
2015
2016
14%
12%
10%
All-Payer
Medicare FFS
8%
6%
4%
2%
Case-Mix Adjusted
Readmissions
CY13 Sept YTD
CY14 Sept YTD
CY15 Sept YTD
CY16 Sept YTD
CY13 - CY16 YTD %
Change
12.91%
12.44%
12.04%
11.49%
Medicare
FFS
13.74%
13.49%
12.95%
12.39%
-11.04%
-9.79%
All-Payer
0%
Note: Based on final data for January 2012 – June 2016, and preliminary data through October 2016.
8
Maryland is reducing readmission rate but only
slightly faster than the nation
18.50%
18.17%
18.00%
17.42%
17.50%
17.00%
16.61%
16.50%
16.47%
15.95%
16.29%
16.00%
15.61%
15.76%
15.50%
15.39%
15.50%
15.42%
15.26%
15.00%
14.50%
14.00%
13.50%
CY2011
CY2012
CY2013
National
9
CY2014
Maryland
CY 2015
CY 2016 YTD Jul
All-Payer Case-Mix Adjusted PPC Rates
CYTD 2013 - CYTD 2016
1.8
New
Model
Start
Date
1.6
Case-Mix Adjusted
Medicare
All-Payer
PPC Rate
FFS
CY13 YTD Through Sep
1.29
1.50
1.4
CY14 YTD Through Sep
0.96
1.06
CY15 YTD Through Sep
0.87
0.99
CY16 YTD Through Sep
0.69
CY16 over CY13 %
Change (YTD through -46.45%
Sep)
1.2
0.76
-49.31%
1.0
0.8
ALL PAYER
MEDICARE FFS
0.6
Linear (ALL
PAYER)
10
Sep-16
Jul-16
May-16
Mar-16
Jan-16
Nov-15
Sep-15
Jul-15
May-15
Mar-15
Jan-15
Nov-14
Sep-14
Jul-14
May-14
Mar-14
Jan-14
Nov-13
Sep-13
Jul-13
May-13
Mar-13
Jan-13
Nov-12
Sep-12
Jul-12
0.4
Value-Based Performance Policies
• All-payer demonstration contract requirements
• HSCRC payment policies
• Exemptions from national value-based payment policies
11
Value-Based Performance Policies
• All-payer demonstration contract language applies to
readmissions and complications
…the state [must] demonstrate that it is implementing a program for
regulated Maryland hospitals and, as applicable, other hospitals in
Maryland that achieves or surpasses the [national] measured results
in terms of patient outcomes and cost savings…
The state must ensure that the aggregate percentage of regulated
revenue at risk for quality programs…is equal to or greater than the
aggregate…at risk under national Medicare quality programs.
•
Value Based Purchasing exemption is an annual process
and follows similar guidelines
12
Updates to HSCRC Value-Based
Payment Policies
13
ICD - 10
ICD - 9
ICD-10 Transition
Period
2013 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
2014 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
2015 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
2016 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
2017 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
2018 Jan - Mar
Apr - Jun
Jul - Sep
Oct - Dec
APRDRG Grouper Version
v 32
v 33
v 34
Readmissions and MHAC
FY 2018
FY 2019
Base Period
Performance Modified
Period for FY Base Period
2018
Adjustment
Performance
Period for
FY 2019
Adjustment
In CY 2017
performance
year, the base
and performance
periods are
coded in ICD-10.
The preference
is to use grouper
version 34
14
Updates to HSCRC Policies
• QBR
– Revenue neutral scaling eliminated. Pre-set scale similar to MHAC
policy
– FY 2018 pre-set scale to be based on performance year 2016 results
– FY 2019 pre-set scale to be determined by commissioners
– Continued emphasis on HCAHPS
• MHAC
– Proposal to remove palliative care exclusion—studying further
– Update thresholds and benchmarks; CY17 performance compared to
October 2015 – September 2016 base period; Grouper version 34
– No state improvement target; single payment scale
15
Updates to HSCRC Policies
• Readmissions
– No significant changes to policy have been discussed
– Delay in setting attainment and improvement targets due to ICD10 anomalies
Policy Year
Improvement Target
FY 2016
6.36%
FY 2017
9.3%
FY 2018
9.5%
FY 2019
??
16
QBR and MHAC Scoring
• Earn between 0-10 points on each metric, which rolls up to a
final score between 0-1
– Better of attainment and improvement
– Threshold – average performance and minimum performance required
to begin earning points
– Benchmark – top (decile) performance for which maximum points are
earned
• Final score of 0 = on every metric the hospital performed
worse than average; 1 = hospital at top performance on every
metric
17
Setting QBR and MHAC Payment Scales
• Three anchor points
– Maximum reward – best score, best possible score, or some other high
point (e.g., 0.8)
– Maximum penalty – lowest score or lowest possible score
– “Break point” – where rewards begin and penalties end; average score
from a current or prior period. This is a value judgment and an
indication of expectations
18
Setting the QBR Payment Scale
FY 16
Permanent
Inpatient
Revenue
HOSPITAL
NAME
A
C
G
I
M
O
S
U
Y
AA
EE
GG
KK
MM
QQ
Statewide
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
74,789,724
126,399,313
156,564,761
415,350,729
238,195,335
237,934,932
190,659,648
69,389,876
135,939,076
906,034,034
214,208,592
136,267,434
75,687,627
190,413,775
165,683,744
$8,730,031,841
QBR FINAL
POINTS
0.07
0.20
0.25
0.31
0.31
0.33
0.36
0.37
0.38
0.39
0.41
0.43
0.45
0.46
0.57
Total Penalties
Total rewards
Range 0.07 to 0.57
Breakpoint at 0.25
% Revenue
Impact
-2.00%
-0.67%
-0.11%
0.18%
0.18%
0.25%
0.36%
0.39%
0.43%
0.46%
0.54%
0.61%
0.68%
0.71%
1.00%
$ Revenue Impact
-$1,495,794
-$846,875
-$172,221
$747,631
$428,752
$594,837
$686,375
$270,621
$584,538
$4,167,757
$1,156,726
$831,231
$514,676
$1,351,938
$1,656,837
$26,366,722
-5,389,617
32,448,031
Range 0.07 to 0.57
Breakpoint at 0.37
% Revenue
Impact
-2.00%
-1.13%
-0.80%
-0.40%
-0.40%
-0.27%
-0.07%
0.00%
0.05%
0.10%
0.20%
0.30%
0.40%
0.45%
1.00%
$ Revenue Impact
-$1,495,794
-$1,432,526
-$1,252,518
-$1,661,403
-$952,781
-$634,493
-$127,106
$0
$67,970
$906,034
$428,417
$408,802
$302,751
$856,862
$1,656,837
-$10,575,223
-20,503,119
10,619,589
19
Modeling of QBR Scaling Options
• Which scores should be used for maximum rewards and penalties ?
• Which score should be used as cut point to turn from penalty to reward
zones ?
• 80% represents realistic max possible score
• Rewards can be increased in commensurate with higher points
• Increase the maximum reward from 1% to 2% inpatient revenue
RY 19 Scaling Options
Min
Cut Point
Max
Final Scores (max reward 1%)
7%
37%
57%
Full Scale Options Max Reward 2%
Full Score Range
0%
50%
100%
Option 1
0%
40%
80%
Option 2
0%
45%
80%
Note: Modeling based on RY17 Final Scores
20
Statewide Statewide
Penalties Rewards
-$20M
+11M
-49M
-24M
-37M
+1M
+7M
+3M
MHAC Scaling Options
RY 2018 Scale
Final MHAC Score
Scores less
than or equal
to
Scores greater
than or equal
to
0.17
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
-3.00%
-2.74%
-2.29%
-1.85%
-1.41%
-0.97%
-0.53%
-0.09%
0.35%
0.79%
1.24%
1.68%
2.12%
-1.00%
-0.88%
-0.67%
-0.46%
-0.25%
-0.04%
0.00%
0.00%
0.17%
0.33%
0.50%
0.67%
0.83%
0.80
0.00%
1.00%
0.51
No
rewards
0.41
Penalty threshold:
Reward Threshold
21
Below Exceed
State
State
Quality Quality
Target Target
0.50
Option 1: Full Scale
without Neutral Zone
Final MHAC Score
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
Penalty/Reward
threshold:
Revenue
Adjustment
-2.00%
-1.80%
-1.60%
-1.40%
-1.20%
-1.00%
-0.80%
-0.60%
-0.40%
-0.20%
0.00%
0.10%
0.20%
0.30%
0.40%
0.50%
0.60%
0.70%
0.80%
0.90%
1.00%
0.50
Option 2: Full Scale with
Neutral Zone
Final MHAC Score
Revenue
Adjustment
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
-2.00%
-1.78%
-1.56%
-1.33%
-1.11%
-0.89%
-0.67%
-0.44%
-0.22%
0.00%
0.00%
0.00%
0.11%
0.22%
0.33%
0.44%
0.56%
0.67%
0.78%
0.89%
1.00%
Penalty threshold:
Reward Threshold
0.45
0.55
Exemptions from Medicare Value-Based
Programs
• Hospital Acquired Conditions and Readmissions exemptions are
part of the all-payer demonstration…as long as we have a program
of similar scope and risk compared to the nation, and are meeting
annual performance targets
• VBP exemption is an annual process where CMS reviews
Maryland’s annual performance and the HSCRC policy
• Maryland programs measure performance across all payers and
adjust all-payer revenue
– Uniformity in performance metrics across all payers
– Harder to compare Maryland performance to nation when Maryland is
focused on different metrics
22
Maryland QBR Compared to CMS VBP
CMS VBP
Domain
Measure
Maryland QBR
Data Type
Weight
Safety
Domain
Measure
Data Type
Weight
NHSN
All-payer
35%
Safety
CAUTI
CLABSI
SSI – colon
SSI – hysterectomy
C.difficile
MRSA
PC-01
PSI-90
NHSN
All-payer
CAUTI
CLABSI
SSI – colon
SSI – hysterectomy
C.difficile
MRSA
PC-01
PSI-90 (suspended)
20%
Medicare only
Patient
Experience
HSCRC All-payer
Patient
Experience
HCAHPS
Required
reporting
through
independent
vendor
All-payer
25%
Outcome*
HCAHPS
Required reporting
through
independent vendor
All-payer
50%
HSCRC data Allpayer
15%
Outcome
30-day mortality, 3
conditions: AMI, heart
failure, pneumonia
CMS data
Medicare only
25%
Medicare spending per
CMS data
30 day episode
Medicare only
25%
All-cause mortality
Efficiency
*Starting in FY2019, CMS will include two new metrics that measure complication rates up to 90 days following elective
primary total hip arthroplasty and/or total knee arthroplasty. Maryland is not able to calculate all-payer complication rates
that extend beyond the hospital stay.
23
Maryland MHAC Compared to CMS HAC
CMS HAC
Maryland MHAC
Number of
measures
6 (PSI-90, CLABSI, CAUTI, Surgical Site
Infections, MRSA, and C. Difficile infection)
Data source
Medicare claims and chart-abstracted
surveillance data
Attainment compared to hospitals nationally.
58 Potentially Preventable Complications (7 of the
original 65 have been removed due to clinical
concerns)
All-payer HSCRC case-mix data
Scoring
Incentives to
individual
hospitals
Better of attainment and improvement relative to
statewide benchmarks
1% decrease in Medicare payments for lowest In years when state meets its collective HAC
performing 25% of hospitals nationally
reduction goal (RY 17)
 Maximum reward: 1% increase of permanent
inpatient revenue
 Maximum penalty: 1% reduction in permanent
inpatient revenue
In years when state fails to meet its collective HAC
reduction goal:

Maximum penalty: 3% reduction of permanent
inpatient revenue
24
Population Health Metrics and
Measurement
25
Population Health Planning
• CMMI
– Maryland must include population health metrics as part of Care
Redesign amendment to All-Payer Demonstration model and
– All-Payer Demonstration Progression Plan
• DHMH’s Office of Population Health Improvement
– Developed a broad-reaching strategic thought framework and
process to address individual risk factors contributing to poor health,
including social determinants of health. The final document,
Maryland Population Health Improvement Plan: Planning for
Population Health Improvement will be posted
at http://pophealth.dhmh.maryland.gov/Pages/transformation.aspx
– Developing a framework to identify priorities, data sources, potential
metrics and accountability mechanisms that will eventually lead to
inclusion of metrics in HSCRC value-based payment policies
26
Draft Population Health Timeline
Due Date
Description
June 30, 2017
State submits a Population Health Plan to CMS.
August 31, 2017
CMS target date to send comments on the submitted Population Health Plan to the State
(requested within 60 calendar days of receiving the State’s Population Health Plan). State works
with CMS to incorporate CMS comments in the Population Health Plan.
January 1, 2018
State submits to CMS the Value Based Payment Plan (“VBP Plan”).
July 1, 2018
State begins tracking proposed value-based program measures for each hospital.
March 31, 2019
Based on the State’s testing, the State submits any modifications to the VBP Plan to CMS for
review and comment.
May 31, 2019
CMS target date to send comments on the submitted VBP Plan to the State (requested within 60
calendar days of receiving the State’s VBP Plan). State works with CMS to incorporate CMS
comments and modifications in the VBP Plan.
July 1, 2019
State incorporates the VBP Plan Measures into its payment methodologies.
27
MHA’s Recommendations to DHMH
1. Hold providers accountable only for outcomes they can
control
• Focus on risk factors that can be influenced by clinical
interventions, e.g., diabetes and hypertension control
• Involve other sectors such as schools, local health departments,
public policy for behavioral risk factors such as smoking
prevalence and obesity
• Public and private entities address shared goals with
interventions and accountability appropriate to their sphere of
influence in a “layered approach”
28
MHA’s Recommendations to DHMH
2. Include behavioral health and access to care
• Emergency Department visits for behavioral health conditions
have increased over the last three years while ED use for all
other conditions has declined
• While Maryland has one of the lowest rates of uninsured,
access to a regular source of primary care is critical to
managing chronic health conditions
29
Example: Layered Measurement Approach and Varying
Accountability in Public Policy and Health Care
Population-based Measurement
• Several policies under development involve measuring individual
hospitals’ impact on a population. Different methods of attributing
individuals and costs may be used depending on the policy goal
– Total cost of care (TCOC) monitoring, e.g., as one of the annual update
conditions (HSCRC’s zip code approach)
– TCOC guardrail, e.g., care redesign activities that include access to detailed
Medicare data require a TCOC guardrail (approach not yet defined)
– HSCRC “geographic model” an extension of GBR to include non-hospital
costs
– Payment policy related to efficiency, e.g., value-based policy, eligibility for
capital funds
• Attribute some or all Medicare beneficiaries
• Consider combination of different approaches, including beneficiary level
attribution, geographic attribution
31
Population-based Measurement
• Total cost of care
– Assigns or attributes most or all beneficiaries to hospitals
– Account for a significant portion of total Medicare costs
– Inherent differences in population’s disease prevalence and
community resources could be addressed by
• Comparing trend in spending per beneficiary
• Risk adjustment
• Beneficiary level analysis is important to understand patterns of use to
inform this approach and to identify potential issues that need to be
addressed in a geographic approach
• However, for some beneficiaries, it may not be clear which hospital has
the lead on managing their care and a geographic model or shared
accountability may be needed
32
Population-based Measurement
• Criteria to assign beneficiaries could include
– Plurality of hospital care
– Total charges
– Types of service, e.g., inpatient admissions, observation, clinic and ED
visits are more likely than diagnostic imaging services to indicate the
hospital that is managing care
– Beneficiary residence
– Physician E & M visits and physician referral patterns
– Combination of factors
• In most cases, the hospital with the plurality of visits is also the hospital with
the highest charges, and it appears that a high percentage of beneficiaries who
use hospital services are clearly linked to a single hospital or system
33
Other Areas Still to be Explored
• Non-hospital utilization
– Post-acute care following a hospital discharge
– Patterns of use and potential linkages to hospital for hospice and
home health
– Non-hospital Part B utilization among high utilizers of hospital
care
• Beneficiaries with little utilization
– Non-hospital Part B utilization only
– No utilization
• Stability of patterns over time
• Comparison to ACO attribution
34
Traci La Valle is a Vice President at the Maryland Hospital Association where she advocates for
Maryland's hospitals, health systems, communities, and patients primarily before state regulatory
bodies. In her role, she works to ensure fair and reasonable hospital payment policies that provide
appropriate incentives to improve quality and reduce avoidable costs. In her years at MHA, she
has held progressively responsible roles covering a range of issues that affect Maryland hospital
finances. Traci has a Master of Public Health and a Certificate in Health Finance and Management
from Johns Hopkins School of Public Health, and a Bachelor of Science in Physical Therapy from
Temple University.
Traci La Valle
Maryland Hospital Association
6820 Deerpath Road
Elkridge, MD 21075
410-540-5087
[email protected]
35
Maryland Value-Based Performance Policies
What we’ve achieved and where we’re headed
Traci La Valle
Vice President
Maryland Hospital Association
January 27, 2017