Nursing Home ValueBased Purchasing Demonstration - PowerPoint PPT Presentation


PPT – Nursing Home ValueBased Purchasing Demonstration PowerPoint presentation | free to view - id: 552b3-ZDc1Z


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

Nursing Home ValueBased Purchasing Demonstration


Nursing Home Value-Based Purchasing Demonstration. Ron Lambert CMS April 16, 2007. OUTLINE ... Sum the values for each nursing home. Rank homes from lowest to ... – PowerPoint PPT presentation

Number of Views:32
Avg rating:3.0/5.0
Slides: 32
Provided by: CMS175


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Nursing Home ValueBased Purchasing Demonstration

Nursing Home Value-Based Purchasing Demonstration
  • Ron Lambert CMS April 16,

  • Background
  • Demonstration design
  • Performance measures
  • Scoring rules
  • Measuring Medicare program savings
  • Allocating the performance pool
  • Solicitation

  • Purpose To test the pay-for-performance
    concept in the nursing home setting prior to
    national implementation
  • Objective To improve the quality of care
    furnished to all Medicare beneficiaries in
    nursing homes
  • Approach To provide financial incentives to
    nursing homes to achieve high standards for
    providing quality of care
  • Implementation contractor Abt Associates and
    University of Colorado Health Sciences Center

Basic Approach
  • CMS will assess the performance of nursing homes
    based on selected quality measures
  • Four domains
  • Nurse staffing
  • Appropriate hospitalizations
  • MDS Outcomes
  • Survey deficiencies
  • CMS will then make additional payments to those
    nursing homes that achieve a higher performance
    level based on the measures within these domains.

  • CMS plans to conduct a 3-year demonstration
    beginning in 2008
  • Two-stage solicitation
  • CMS will select 4 or 5 States
  • States must expect to have at least 100 nursing
    homes willing to apply
  • CMS will solicit nursing homes within the
    selected States
  • The demonstration will include free-standing
    homes and hospital-based facilities

Design Characteristics
  • Demonstration will include all Medicare
    beneficiaries residing in nursing homes
  • Design must account for differences in short and
    long stay populations
  • Demonstration must be budget neutral with respect
    to Medicare
  • Improvements in quality must result in a
    savings pool that can be used to fund the
  • CMS wants to conduct the project in cooperation
    with States

Operational Considerations
  • Select quality measures that are meaningful,
    verifiable, and feasible to collect
  • Use existing data to the extent possible
  • New measures may require extensive development
  • Minimizes reporting burden
  • Measure performance and make payment awards on an
    annual basis
  • Implement performance measures and payment pools
    separately for each State
  • Performance may vary across geographic areas

Budget Neutrality
  • Within each State, CMS will randomly assign
    volunteer nursing homes to experimental
    (demonstration) and control groups
  • About 50 nursing homes will be assigned to each
  • The size of the States payment pool will be
    determined based on the estimated Medicare
    program savings achieved by the demonstration
    homes in each State
  • We anticipate that higher quality of care will
    result in fewer inappropriate hospitalizations
  • Reductions in Part A hospital stays and
    subsequent Part A nursing home stays should
    result in savings to Medicare

Performance Measures
  • CMS has selected a set of quality measures upon
    which quality of care in nursing homes will be
  • Each year of the demonstration, CMS will
    calculate a score for each nursing home based on
    its performance on these measures
  • CMS will also consider developmental measures for
    possible inclusion in year 2 of the demonstration
  • Staff immunization rates
  • Use of experience of care surveys

  • There is strong evidence that low nurse staffing
    and high staff turnover compromise the quality of
    care of nursing home residents
  • Measures
  • RN hours per resident day
  • Total nursing hours per resident day (RN, LPN,
    nurse aide)
  • Turnover for nursing staff
  • We will calculate these measures from payroll
    data submitted by nursing homes
  • Required data elements will include employee
    start date, job category, number of production
    hours worked

Appropriate Hospitalizations
  • Studies suggest that a substantial portion of
    hospital admissions of nursing home residents are
    inappropriate, and can be avoided through careful
    management of conditions
  • We are considering including the following
    conditions CHF, electrolyte imbalance,
    respiratory disease, sepsis, urinary tract

Hospitalizations (continued)
  • There are separate measures for short and long
    stay residents
  • Rate of potentially avoidable hospitalizations
    for short stay residents
  • Rate of potentially avoidable hospitalizations
    for long stay residents
  • Hospitalization rate is defined as number of
    hospitalizations for an included condition
    divided by the number of resident days

MDS Outcome Measures
  • Use of MDS measures aligns payment incentives
    with achievement of better outcomes
  • We will use a subset of MDS-based measures that
    have been found to be valid and reliable
  • Underlying assumption is that the variation of
    outcome measures across nursing homes is affected
    by the varying quality of care, and that these
    measures can be impacted by changing the care
    practice in a nursing home

MDS Outcome Measures Long Stay Residents
  • Measures
  • Percent of residents whose need for help with
    daily activities has increased
  • Percent of residents whose ability to move in and
    around their room got worse
  • Percent of high-risk residents who have pressure
  • Percent of residents who have had a catheter left
    in their bladder
  • Percent of residents who were physically

MDS Outcome Measures Short Stay Residents
  • Measures
  • Percent of residents with improving level of ADL
  • Percent of residents who improve status on
    mid-loss ADL functioning
  • Percent of residents who fail to improve bladder

Survey Deficiencies
  • Certification surveys provide a broad perspective
    of the quality of care furnished by nursing homes
  • Surveys will be used in two ways
  • Qualification
  • Nursing homes will be ineligible for a
    performance payment if they were cited for
    substandard quality of care during the
    demonstration year
  • Measure
  • Deficiencies on standard and complaint surveys
    will be assigned values based on their scope and
  • Nursing homes will be ranked within each State
    according to their values

Developmental Measures
  • CMS will evaluate these measures for inclusion in
    year 2 of the demonstration
  • CMS will collect these measures via a data
    collection tool
  • Baseline information will be collected from
    applicant nursing homes
  • Measures that are under consideration include
  • Staff immunization rates
  • Use of experience of care surveys

Adjustments for Case-Mix
  • Staffing
  • Use RUG-III nursing index to adjust total nursing
  • For RN measure, use separate scoring for
    hospital-based and free-standing nursing homes
  • Hospitalizations
  • We will develop a model to risk adjust
    hospitalization rates
  • MDS Outcomes
  • Long-stay measures use the methodology used for
    reporting on nursing home compare
  • Short-stay measures use risk adjustment and
    exclusion methodology used in the Abt validation

Point Allocation
  • Points will be allocated to each of the domains
    as follows
  • Staffing 30 points
  • 10 points each measure
  • Hospitalizations 30 points
  • MDS Outcomes 20 points
  • Equal points for each measure
  • Survey Deficiencies 20 points

Relative Ranking
  • For each measure, we will rank nursing homes
    relative to other nursing homes in the State
  • Points will be assigned across the entire range
    of the ranked distribution
  • This avoids thresholds where small differences in
    performance could yield large differences in
  • Exception there will be a maximum threshold for
    the hospitalization domain

Scoring Example Survey
  • Values are assigned to each deficiency as

  • Scope
  • Severity
    isolated pattern
  • Immediate jeopardy to resident
    J K
  • health or safety
    (50) (100)
  • Actual harm that is not immediate
    G H
  • Jeopardy
    (10) (20)
  • No actual harm with potential for
    D E
  • more than minimal harm that is not
    (2) (4)
  • immediate jeopardy
  • No actual harm with potential for
    A B
  • more than minimal harm
    (0) (0)

Scoring Example (continued)
  • Sum the values for each nursing home
  • Rank homes from lowest to highest values
  • Nursing home with lowest value (fewest
    deficiencies) gets score of 20 nursing home with
    highest value gets score of zero
  • Scores for other nursing homes are based on
    relative ranking

Estimating the Savings Pool
  • Medicare program savings will be determined each
    year for each State
  • The savings estimate will be based on Medicare
    expenditures per resident day
  • Include inpatient, SNF, outpatient, hospice, and
    physician/supplier claims.
  • Include claims with date of service within the
    period that the beneficiary resided in the
    nursing home and Medicare-covered services within
    3 days after nursing home discharge
  • For inpatient hospital stays that begin within
    the three day window, the entire hospital stay
    will be included

Basic Methodology
  • Compare change in expenditures for residents of
    demonstration nursing homes (experimental group)
    to the change in expenditures for residents of
    nursing homes in the control group
  • Medicare expenditures will be risk adjusted
  • CMS will develop an appropriate risk adjustment

Methodology (continued)
  • Step 1 Calculate risk-adjusted Medicare
    expenditures for demonstration and control
    beneficiaries for the baseline and demonstration
  • Step 2 Calculate expected Medicare expenditures
    for demonstration beneficiaries
  • Estimate the expenditures for beneficiaries in
    demonstration homes in the absence of the
  • This depends on the baseline expenditures for the
    demonstration group and the change in
    expenditures for the control group
  • Step 3 Calculate Medicare savings as the
    difference between actual and expected Medicare
    expenditures per day

Illustration Savings Calculation

  • demonstration group
    control group

  • base year year 1
    base year year 1
  • Risk adjusted
  • Medicare expenditures 10,000,000
    12,000,000 10,800,000
  • Total resident days
    200,000 230,707
    180,000 200,000
  • Medicare expenditures
  • Per day
    50.00 52.00
    60.00 64,80
  • Change in Medicare
    (52 - 50) / 50
    (64.80 - 60) / 60
  • Expenditures
  • Expected Medicare
  • Expenditures - Year 1
  • Medicare savings per day
    (54.00 - 52.00) 2.00

Allocating the Pool - Principals
  • Reward high performing nursing homes
  • Encourage improvement in nursing homes that may
    not have good quality initially
  • Improvers must also meet a minimum performance
  • Reward nursing homes based on overall performance
  • Intent reward homes that provide overall high
    quality care rather than those that excel in
    certain areas

Allocating the Pool
  • A nursing home will qualify for a performance
    payment if it is either
  • in the top 20 percent in overall performance or
  • In the top 20 percent in terms of improvement
  • Those in the top decile will receive more than
    those in the second decile
  • Nursing homes with both high performance and
    improvement will be rewarded for either
    performance or improvement, not both
  • Homes will receive the higher of the two payments

Allocating (continued)
  • Savings pool would be split evenly between
    performers and improvers
  • Performance payments would be weighted by nursing
    homes average number of Medicare residents per

  • CMS will solicit host States via an invitation
    letter to State Medicaid Directors
  • Status letter is in clearance
  • Once the States are selected, CMS will solicit
    volunteer nursing homes
  • Applicants will be required to submit baseline
    information on staffing, resident census,
    developmental measures

  • The Nursing Home Value-Based Purchasing
    Demonstration has several noteworthy features
  • It encompasses all Medicare beneficiaries
  • The performance measures reflect various
    dimensions of quality
  • By rewarding both performers and improvers, all
    demonstration nursing homes can potentially
    qualify for a performance award
  • If the demonstration is successful, NHVBP could
    be implemented nationally